User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability

Visualizing Clinical Data Retrieval and Curation in Multimodal Healthcare AI Research: A Technical Note on RIL-workflow

Curating and integrating data from sources are bottlenecks to procuring robust training datasets for artificial intelligence (AI) models in healthcare. While numerous applications can process discrete types of clinical data, it is still time-consuming to integrate heterogenous data types. Therefore, there exists a need for more efficient retrieval and storage of curated patient data from dissimilar sources, such as biobanks, health records, and sensors. We describe a customizable, modular data retrieval application (RIL-workflow), which integrates clinical notes, images, and prescription data, and show its feasibility applied to research at our institution. It uses the workflow automation platform Camunda (Camunda Services GmbH, Berlin, Germany) to collect internal data from Fast Healthcare Interoperability Resources (FHIR) and Digital Imaging and Communications in Medicine (DICOM) sources. Using the web-based graphical user interface (GUI), the workflow runs tasks to completion according to visual representation, retrieving and storing results for patients meeting study inclusion criteria while segregating errors for human review. We showcase RIL-workflow with its library of ready-to-use modules, enabling researchers to specify human input or automation at fixed steps. We validated our workflow by demonstrating its capability to aggregate, curate, and handle errors related to data from multiple sources to generate a multimodal database for clinical AI research. Further, we solicited user feedback to highlight the pros and cons associated with RIL-workflow. The source code is available at github.com/magnooj/RIL-workflow.

MR Neurography and Quantitative Muscle MRI of Parsonage Turner Syndrome Involving the Long Thoracic Nerve

BACKGROUND

Parsonage-Turner syndrome (PTS) is characterized by severe, acute upper extremity pain and subsequent paresis and most commonly involves the long thoracic nerve (LTN). While MR neurography (MRN) can detect LTN hourglass-like constrictions (HGCs), quantitative muscle MRI (qMRI) can quantify serratus anterior muscle (SAM) neurogenic changes.

PURPOSE/HYPOTHESIS

1) To characterize qMRI findings in LTN-involved PTS. 2) To investigate associations between qMRI and clinical assessments of HGCs/electromyography (EMG).

STUDY TYPE

Prospective.

POPULATION

30 PTS subjects (25 M/5 F, mean/range age = 39/15-67 years) with LTN involvement who underwent bilateral chest wall qMRI and unilateral brachial plexus MRN.

FIELD STRENGTH/SEQUENCES

3.0 Tesla/multiecho spin-echo T2-mapping, diffusion-weighted echo-planar-imaging, multiecho gradient echo.

ASSESSMENT

qMRI was performed to obtain T2, muscle diameter fat fraction (FF), and cross-sectional area of the SAM. Clinical reports of MRN and EMG were obtained; from MRN, the number of HGCs; from EMG, SAM measurements of motor unit recruitment levels, fibrillations, and positive sharp waves. qMRI/MRN were performed within 90 days of EMG. EMG was performed on average 185 days from symptom onset (all ≥2 weeks from symptom onset) and 5 days preceding MRI.

STATISTICAL TESTS

Paired t-tests were used to compare qMRI measures in the affected SAM versus the contralateral, unaffected side (P < 0.05 deemed statistically significant). Kendall's tau was used to determine associations between qMRI against HGCs and EMG.

RESULTS

Relative to the unaffected SAM, the affected SAM had increased T2 (50.42 ± 6.62 vs. 39.09 ± 4.23 msec) and FF (8.45 ± 9.69 vs. 4.03% ± 1.97%), and decreased muscle diameter (74.26 ± 21.54 vs. 88.73 ± 17.61 μm) and cross-sectional area (9.21 ± 3.75 vs. 16.77 ± 6.40 mm2). There were weak to negligible associations (tau = -0.229 to <0.001, P = 0.054-1.00) between individual qMRI biomarkers and clinical assessments of HGCs and EMG.

DATA CONCLUSION

qMRI changes in the SAM were observed in subjects with PTS involving the LTN.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY: Stage 1.

Deep learning for colorectal cancer detection in contrast-enhanced CT without bowel preparation: a retrospective, multicentre study

BACKGROUND

Contrast-enhanced CT scans provide a means to detect unsuspected colorectal cancer. However, colorectal cancers in contrast-enhanced CT without bowel preparation may elude detection by radiologists. We aimed to develop a deep learning (DL) model for accurate detection of colorectal cancer, and evaluate whether it could improve the detection performance of radiologists.

METHODS

We developed a DL model using a manually annotated dataset (1196 cancer vs 1034 normal). The DL model was tested using an internal test set (98 vs 115), two external test sets (202 vs 265 in 1, and 252 vs 481 in 2), and a real-world test set (53 vs 1524). We compared the detection performance of the DL model with radiologists, and evaluated its capacity to enhance radiologists' detection performance.

FINDINGS

In the four test sets, the DL model had the area under the receiver operating characteristic curves (AUCs) ranging between 0.957 and 0.994. In both the internal test set and external test set 1, the DL model yielded higher accuracy than that of radiologists (97.2% vs 86.0%, p < 0.0001; 94.9% vs 85.3%, p < 0.0001), and significantly improved the accuracy of radiologists (93.4% vs 86.0%, p < 0.0001; 93.6% vs 85.3%, p < 0.0001). In the real-world test set, the DL model delivered sensitivity comparable to that of radiologists who had been informed about clinical indications for most cancer cases (94.3% vs 96.2%, p > 0.99), and it detected 2 cases that had been missed by radiologists.

INTERPRETATION

The developed DL model can accurately detect colorectal cancer and improve radiologists' detection performance, showing its potential as an effective computer-aided detection tool.

FUNDING

This study was supported by National Science Fund for Distinguished Young Scholars of China (No. 81925023); Regional Innovation and Development Joint Fund of National Natural Science Foundation of China (No. U22A20345); National Natural Science Foundation of China (No. 82072090 and No. 82371954); Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application (No. 2022B1212010011); High-level Hospital Construction Project (No. DFJHBF202105).

90Y post-radioembolization clinical assessment with whole-body Biograph Vision Quadra PET/CT: image quality, tumor, liver and lung dosimetry

PURPOSE

Evaluation of 90Y liver radioembolization post-treatment clinical data using a whole-body Biograph Vision Quadra PET/CT to investigate the potential of protocol optimization in terms of scan time and dosimetry.

METHODS

17 patients with hepatocellular carcinoma with median (IQR) injected activity 2393 (1348-3298) MBq were included. Pre-treatment dosimetry plan was based on 99mTc-MAA SPECT/CT with Simplicit90Y™ and post-treatment validation with Quadra using Simplicit90Y™ and HERMIA independently. Regarding the image analysis, mean and peak SNR, the coefficient of variation (COV) and lesion-to-background ratio (LBR) were evaluated. For the post-treatment dosimetry validation, the mean tumor, whole liver and lung absorbed dose evaluation was performed using Simplicit90Y and HERMES. Images were reconstructed with 20-, 15-, 10-, 5- and 1- min sinograms with 2, 4, 6 and 8 iterations. Wilcoxon signed rank test was used to show statistical significance (p < 0.05).

RESULTS

There was no difference of statistical significance between 20- and 5- min reconstructed times for the peak SNR, COV and LBR. In addition, there was no difference of statistical significance between 20- and 1- min reconstructed times for all dosimetry metrics. Lung dosimetry showed consistently lower values than the expected. Tumor absorbed dose based on Simplicit90Y™ was similar to the expected while HERMES consistently underestimated significantly the measured tumor absorbed dose. Finally, there was no difference of statistical significance between expected and measured tumor, whole liver and lung dose for all reconstruction times.

CONCLUSION

In this study we evaluated, in terms of image quality and dosimetry, whole-body PET clinical images of patients after having been treated with 90Y microspheres radioembolization for liver cancer. Compared to the 20-min standard scan, the simulated 5-min reconstructed images provided equal image peak SNR and noise behavior, while performing also similarly for post-treatment dosimetry of tumor, whole liver and lung absorbed doses.

EfficientNet-Based System for Detecting EGFR-Mutant Status and Predicting Prognosis of Tyrosine Kinase Inhibitors in Patients with NSCLC

We aimed to develop and validate a deep learning-based system using pre-therapy computed tomography (CT) images to detect epidermal growth factor receptor (EGFR)-mutant status in patients with non-small cell lung cancer (NSCLC) and predict the prognosis of advanced-stage patients with EGFR mutations treated with EGFR tyrosine kinase inhibitors (TKI). This retrospective, multicenter study included 485 patients with NSCLC from four hospitals. Of them, 339 patients from three centers were included in the training dataset to develop an EfficientNetV2-L-based model (EME) for predicting EGFR-mutant status, and the remaining patients were assigned to an independent test dataset. EME semantic features were extracted to construct an EME-prognostic model to stratify the prognosis of EGFR-mutant NSCLC patients receiving EGFR-TKI. A comparison of EME and radiomics was conducted. Additionally, we included patients from The Cancer Genome Atlas lung adenocarcinoma dataset with both CT images and RNA sequencing data to explore the biological associations between EME score and EGFR-related biological processes. EME obtained an area under the curve (AUC) of 0.907 (95% CI 0.840-0.926) on the test dataset, superior to the radiomics model (P = 0.007). The EME and radiomics fusion model showed better (AUC, 0.941) but not significantly increased performance (P = 0.895) compared with EME. In prognostic stratification, the EME-prognostic model achieved the best performance (C-index, 0.711). Moreover, the EME-prognostic score showed strong associations with biological pathways related to EGFR expression and EGFR-TKI efficacy. EME demonstrated a non-invasive and biologically interpretable approach to predict EGFR status, stratify survival prognosis, and correlate biological pathways in patients with NSCLC.

Impact of 4D-Flow CMR Parameters on Functional Evaluation of Fontan Circulation

We sought to evaluate the potential clinical role of 4D-flow cardiac magnetic resonance (CMR)-derived energetics and flow parameters in a cohort of patients' post-Fontan palliation. In patients with Fontan circulation who underwent 4D-Flow CMR, streamlines distribution was evaluated, as well a 4D-flow CMR-derived energetics parameters as kinetic energy (KE) and energy loss (EL) normalized by volume. EL/KE index as a marker of flow efficiency was also calculated. Cardiopulmonary exercise test (CPET) was also performed in a subgroup of patients. The population study included 55 patients (mean age 22 ± 11 years). The analysis of the streamlines revealed a preferential distribution of the right superior vena cava flow for the right pulmonary artery (62.5 ± 35.4%) and a mild preferential flow for the left pulmonary artery (52.3 ± 40.6%) of the inferior vena cave-pulmonary arteries (IVC-PA) conduit. Patients with heart failure (HF) presented lower IVC/PA-conduit flow (0.75 ± 0.5 vs 1.3 ± 0.5 l/min/m2, p = 0.004) and a higher mean flow-jet angle of the IVC-PA conduit (39.2 ± 22.8 vs 15.2 ± 8.9, p < 0.001) than the remaining patients. EL/KE index correlates inversely with VO2/kg/min: R: - 0.45, p = 0.01 peak, minute ventilation (VE) R: - 0.466, p < 0.01, maximal voluntary ventilation: R:0.44, p = 0.001 and positively with the physiological dead space to the tidal volume ratio (VD/VT) peak: R: 0.58, p < 0.01. From our data, lower blood flow in IVC/PA conduit and eccentric flow was associated with HF whereas higher EL/KE index was associated with reduced functional capacity and impaired lung function. Larger studies are needed to confirm our results and to further improve the prognostic role of the 4D-Flow CMR in this challenging population.

Post-mortem changes of anisotropic mechanical properties in the porcine brain assessed by MR elastography

Knowledge of the mechanical properties of brain tissue in vivo is essential to understanding the mechanisms underlying traumatic brain injury (TBI) and to creating accurate computational models of TBI and neurosurgical simulation. Brain white matter, which is composed of aligned, myelinated, axonal fibers, is structurally anisotropic. White matter in vivo also exhibits mechanical anisotropy, as measured by magnetic resonance elastography (MRE), but measurements of anisotropy obtained by mechanical testing of white matter ex vivo have been inconsistent. The minipig has a gyrencephalic brain with similar white matter and gray matter proportions to humans and therefore provides a relevant model for human brain mechanics. In this study, we compare estimates of anisotropic mechanical properties of the minipig brain obtained by identical, non-invasive methods in the live (in vivo) and dead animals (in situ). To do so, we combine wave displacement fields from MRE and fiber directions derived from diffusion tensor imaging (DTI) with a finite element-based, transversely-isotropic nonlinear inversion (TI-NLI) algorithm. Maps of anisotropic mechanical properties in the minipig brain were generated for each animal alive and at specific times post-mortem. These maps show that white matter is stiffer, more dissipative, and more anisotropic than gray matter when the minipig is alive, but that these differences largely disappear post-mortem, with the exception of tensile anisotropy. Overall, brain tissue becomes stiffer, less dissipative, and less mechanically anisotropic post-mortem. These findings emphasize the importance of testing brain tissue properties in vivo.

Statement of Significance

In this study, MRE and DTI in the minipig were combined to estimate, for the first time, anisotropic mechanical properties in the living brain and in the same brain after death. Significant differences were observed in the anisotropic behavior of brain tissue post-mortem. These results demonstrate the importance of measuring brain tissue properties in vivo as well as ex vivo, and provide new quantitative data for the development of computational models of brain biomechanics.

DCES-PA: Deformation-controllable elastic shape model for 3D bone proliferation analysis using hand HR-pQCT images

Bone proliferation is an important pathological feature of inflammatory rheumatic diseases. Although recent advance in high-resolution peripheral quantitative computed tomography (HR-pQCT) enables physicians to study microarchitectures, physicians' annotation of proliferation suffers from slice inconsistency and subjective variations. Also, there are only few effective automatic or semi-automatic tools for proliferation detection. In this study, by integrating pathological knowledge of proliferation formation with the advancement of statistical shape analysis theory, we present an unsupervised method, named Deformation-Controllable Elastic Shape model, for 3D bone Proliferation Analysis (DCES-PA). Unlike previous shape analysis methods that directly regularize the smoothness of the displacement field, DCES-PA regularizes the first and second-order derivative of the displacement field and decomposes these vector fields according to different deformations. For the first-order elastic metric, DCES-PA orthogonally decomposes the first-order derivative of the displacement field by shearing, scaling and bending deformation, and then penalize deformations triggering proliferation formation. For the second-order elastic metric, DCES-PA encodes both intrinsic and extrinsic surface curvatures into the second-order derivative of the displacement field to control the generation of high-curvature regions. By integrating the elastic shape metric with the varifold distances, DCES-PA achieves correspondence-free shape analysis. Extensive experiments on both simulated and real clinical datasets demonstrate that DCES-PA not only shows an improved accuracy than other state-of-the-art shape-based methods applied to proliferation analysis but also produces highly sensitive proliferation annotations to assist physicians in proliferation analysis.

Hippocampal subfields and thalamic nuclei associations with clinical outcomes in multiple sclerosis: An ultrahigh field MRI study

BACKGROUND

Previous studies have shown that thalamic and hippocampal neurodegeneration is associated with clinical decline in Multiple Sclerosis (MS). However, contributions of the specific thalamic nuclei and hippocampal subfields require further examination.

OBJECTIVE

Using 7 Tesla (7T) magnetic resonance imaging (MRI), we investigated the cross-sectional associations between functionally grouped thalamic nuclei and hippocampal subfields volumes and T1 relaxation times (T1-RT) and subsequent clinical outcomes in MS.

METHODS

High-resolution T1-weighted and T2-weighted images were acquired at 7T (n=31), preprocessed, and segmented using the Thalamus Optimized Multi Atlas Segmentation (THOMAS, for thalamic nuclei) and the Automatic Segmentation of Hippocampal Subfields (ASHS, for hippocampal subfields) packages. We calculated Pearson correlations between hippocampal subfields and thalamic nuclei volumes and T1-RT and subsequent multi-modal rater-determined and patient-reported clinical outcomes (∼2.5 years after imaging acquisition), correcting for confounders and multiple tests.

RESULTS

Smaller volume bilaterally in the anterior thalamus region correlated with worse performance in gait function, as measured by the Patient Determined Disease Steps (PDDS). Additionally, larger volume in most functional groups of thalamic nuclei correlated with better visual information processing and cognitive function, as measured by the Symbol Digit Modalities Test (SDMT). In bilateral medial and left posterior thalamic regions, there was an inverse association between volumes and T1-RT, potentially indicating higher tissue degeneration in these regions. We also observed marginal associations between the right hippocampal subfields (both volumes and T1-RT) and subsequent clinical outcomes, though they did not survive correction for multiple testing.

CONCLUSION

Ultrahigh field MRI identified markers of structural damage in the thalamic nuclei associated with subsequently worse clinical outcomes in individuals with MS. Longitudinal studies will enable better understanding of the role of microstructural integrity in these brain regions in influencing MS outcomes.

Longitudinal microstructural alterations surrounding subcortical ischemic stroke lesions detected by free-water imaging

In this study we explore the spatio-temporal trajectory and clinical relevance of microstructural white matter changes within and beyond subcortical stroke lesions detected by free-water imaging. Twenty-seven patients with subcortical infarct with mean age of 66.73 (SD 11.57) and median initial NIHSS score of 4 (IQR 3-7) received diffusion MRI 3-5 days, 1 month, 3 months, and 12 months after symptom-onset. Extracellular free-water and fractional anisotropy of the tissue (FAT) were averaged within stroke lesions and the surrounding tissue. Linear models showed increased free-water and decreased FAT in the white matter of patients with subcortical stroke (lesion [free-water/FAT, mean relative difference in %, ipsilesional vs. contralesional hemisphere at 3-5 days, 1 month, 3 months, and 12 months after symptom-onset]: +41/-34, +111/-37, +208/-26, +251/-18; perilesional tissue [range in %]: +[5-24]/-[0.2-7], +[2-20]/-[3-16], +[5-43]/-[2-16], +[10-110]/-[2-12]). Microstructural changes were most prominent within the lesion and gradually became less pronounced with increasing distance from the lesion. While free-water elevations continuously increased over time and peaked after 12 months, FAT decreases were most evident 1 month post-stroke, gradually returning to baseline values thereafter. Higher perilesional free-water and higher lesional FAT at baseline were correlated with greater reductions in lesion size (rho = -0.51, p = .03) in unadjusted analyses only, while there were no associations with clinical measures. In summary, we find a characteristic spatio-temporal pattern of extracellular and cellular alterations beyond subcortical stroke lesions, indicating a dynamic parenchymal response to ischemia characterized by vasogenic edema, cellular damage, and white matter atrophy.

Perfusion of hepatocellular carcinomas measured by diffusion-derived vessel density biomarker: Higher hepatocellular carcinoma perfusion than earlier intravoxel incoherent motion reports

Diffusion-derived vessel density (DVDD) is a physiological surrogate of the area of microvessels per unit tissue area. DDVD is calculated according to DDVD(b0b2) = Sb0/ROIarea0 - Sb2/ROIarea2, where Sb0 and Sb2 refer to the liver signal when b is 0 or 2 s/mm2. Pathohistological studies and contrast-enhanced CT/MRI data showed higher blood volume in hepatocellular carcinoma (HCC) relative to native liver tissue. With intravoxel incoherent motion (IVIM) imaging, most authors paradoxically reported a decreased perfusion fraction of HCC relative to the adjacent liver. This study applied DDVD to assess the perfusion of HCC. MRI was performed with a 3.0-T magnet. Diffusion-weighted images with b-values of 0 and 2 s/mm2 were acquired in 72 HCC patients. Thirty-two patients had microvascular invasion (MVI(+)) and 40 patients did not have microvascular invasion (MVI(-)). Fifty-eight patients had Edmondson-Steiner grade I or II HCC, and 14 patients had Edmondson-Steiner grade III or IV HCC. DDVD measurement was conducted on the axial slice that showed the largest HCC size. DDVD(b0b2) T/L = HCC DDVD(b0b2)/liver DDVD(b0b2). DDVD(b0b2) T/L median (95% confidence interval) of all HCCs was 2.942 (2.419-3.522), of MVI(-) HCCs was 2.699 (2.030-3.522), of MVI(+) HCCs was 2.988 (2.423-3.990), of Edmondson-Steiner grade I/II HCCs was 2.873 (2.277-3.465), and of Edmondson-Steiner grade III/IV HCCs was 3.403 (2.008-4.485). DDVD(b0b2) T/L approximately agrees with contrast agent dynamically enhanced CT/MRI literature data, whereas it differs from earlier IVIM study results, where HCC perfusion fraction was paradoxically lower relative to native liver tissue. A weak trend was noted with MIV(+) HCCs had a higher DDVD(b0b2) T/L than that of MVI(-) HCCs, and a weak trend was noted with the poorly differentiated group of HCCs (Edmondson-Steiner grade III and IV) had a higher DDVD(b0b2) T/L than that of the better differentiated group of HCCs (Edmondson-Steiner grade I and II).

A prognostic neural epigenetic signature in high-grade glioma

Neural-tumor interactions drive glioma growth as evidenced in preclinical models, but clinical validation is limited. We present an epigenetically defined neural signature of glioblastoma that independently predicts patients' survival. We use reference signatures of neural cells to deconvolve tumor DNA and classify samples into low- or high-neural tumors. High-neural glioblastomas exhibit hypomethylated CpG sites and upregulation of genes associated with synaptic integration. Single-cell transcriptomic analysis reveals a high abundance of malignant stemcell-like cells in high-neural glioblastoma, primarily of the neural lineage. These cells are further classified as neural-progenitor-cell-like, astrocyte-like and oligodendrocyte-progenitor-like, alongside oligodendrocytes and excitatory neurons. In line with these findings, high-neural glioblastoma cells engender neuron-to-glioma synapse formation in vitro and in vivo and show an unfavorable survival after xenografting. In patients, a high-neural signature is associated with decreased overall and progression-free survival. High-neural tumors also exhibit increased functional connectivity in magnetencephalography and resting-state magnet resonance imaging and can be detected via DNA analytes and brain-derived neurotrophic factor in patients' plasma. The prognostic importance of the neural signature was further validated in patients diagnosed with diffuse midline glioma. Our study presents an epigenetically defined malignant neural signature in high-grade gliomas that is prognostically relevant. High-neural gliomas likely require a maximized surgical resection approach for improved outcomes.

Resection of the piriform cortex for temporal lobe epilepsy: a Novel approach on imaging segmentation and surgical application

BACKGROUND

The piriform cortex (PC) occupies both banks of the endorhinal sulcus and has an important role in the pathophysiology of temporal lobe epilepsy (TLE). A recent study showed that resection of more than 50% of PC increased the odds of becoming seizure free by a factor of 16.

OBJECTIVE

We report the feasibility of manual segmentation of PC and application of the Geodesic Information Flows (GIF) algorithm to automated segmentation, to guide resection.

METHODS

Manual segmentation of PC was performed by two blinded independent examiners in 60 patients with TLE (55% Left TLE, 52% female) with a median age of 35 years (IQR, 29-47 years) and 20 controls (60% Women) with a median age of 39.5 years (IQR, 31-49). The GIF algorithm was used to create an automated pipeline for parcellating PC which was used to guide excision as part of temporal lobe resection for TLE.

RESULTS

Right PC was larger in patients and controls. Parcellation of PC was used to guide anterior temporal lobe resection, with subsequent seizure freedom and no visual field or language deficit.

CONCLUSION

Reliable segmentation of PC is feasible and can be applied prospectively to guide neurosurgical resection that increases the chances of a good outcome from temporal lobe resection for TLE.

The Quantitative Evaluation of Automatic Segmentation in Lumbar Magnetic Resonance Images

OBJECTIVE

This study aims to overcome challenges in lumbar spine imaging, particularly lumbar spinal stenosis, by developing an automated segmentation model using advanced techniques. Traditional manual measurement and lesion detection methods are limited by subjectivity and inefficiency. The objective is to create an accurate and automated segmentation model that identifies anatomical structures in lumbar spine magnetic resonance imaging scans.

METHODS

Leveraging a dataset of 539 lumbar spinal stenosis patients, the study utilizes the residual U-Net for semantic segmentation in sagittal and axial lumbar spine magnetic resonance images. The model, trained to recognize specific tissue categories, employs a geometry algorithm for anatomical structure quantification. Validation metrics, like Intersection over Union (IOU) and Dice coefficients, validate the residual U-Net's segmentation accuracy. A novel rotation matrix approach is introduced for detecting bulging discs, assessing dural sac compression, and measuring yellow ligament thickness.

RESULTS

The residual U-Net achieves high precision in segmenting lumbar spine structures, with mean IOU values ranging from 0.82 to 0.93 across various tissue categories and views. The automated quantification system provides measurements for intervertebral disc dimensions, dural sac diameter, yellow ligament thickness, and disc hydration. Consistency between training and testing datasets assures the robustness of automated measurements.

CONCLUSION

Automated lumbar spine segmentation with residual U-Net and deep learning exhibits high precision in identifying anatomical structures, facilitating efficient quantification in lumbar spinal stenosis cases. The introduction of a rotation matrix enhances lesion detection, promising improved diagnostic accuracy, and supporting treatment decisions for lumbar spinal stenosis patients.

Consistent spatial lesion-symptom patterns: A comprehensive analysis using triangulation in lesion-symptom mapping in a cohort of stroke patients

INTRODUCTION

The relationship between brain lesions and stroke outcomes is crucial for advancing patient prognosis and developing effective therapies. Stroke is a leading cause of disability worldwide, and it is important to understand the neurological basis of its varied symptomatology. Lesion-symptom mapping (LSM) methods provide a means to identify brain areas that are strongly associated with specific symptoms. However, inner variations in LSM methods can yield different results. To address this, our study aimed to characterize the lesion-symptom mapping variability using three different LSM methods. Specifically, we sought to determine a lesion symptom core across LSM approaches enhancing the robustness of the analysis and removing potential spatial bias.

MATERIAL & METHODS

A cohort consisting of 35 patients with either right- or left-sided middle cerebral artery strokes were enrolled and evaluated using the NIHSS at 24 h post-stroke. Anatomical T1w MRI scans were also obtained 24 h post-stroke. Lesion masks were segmented manually and three distinctive LSM methods were implemented: ROI correlation-based, univariate, and multivariate approaches.

RESULTS

The results of the LSM analyses showed substantial spatial differences in the extension of each of the three lesion maps. However, upon overlaying all three lesion-symptom maps, a consistent lesion core emerged, corresponding to the territory associated with elevated NIHSS scores. This finding not only enhances the spatial accuracy of the lesion map but also underscores its clinical relevance.

CONCLUSION

This study underscores the significance of exploring complementary LSM approaches to investigate the association between brain lesions and stroke outcomes. By utilizing multiple methods, we can increase the robustness of our results, effectively addressing and neutralizing potential spatial bias introduced by each individual method. Such an approach holds promise for enhancing our understanding of stroke pathophysiology and optimizing patient care strategies.

Does adjunctive fixation in conjunction with miniplate affect condylar position and morphology after mandibular advancement through bilateral sagittal split ramus osteotomy? A retrospective 3-dimensional CT comparative study

To minimize condylar positional and morphological changes after mandibular advancement through bilateral sagittal split ramus osteotomy (BSSRO), surgeons add either a bicortical screw or a two-hole plate distal to the conventional single miniplate. Since there have been no previous studies investigating the effect of this combination, our study aimed to evaluate the short- and long-term effects of these adjunctive fixation methods (AFM) on condylar positional and morphological changes after mandibular advancement through BSSRO. This retrospective cohort study included consecutive patients with retruded mandibles who were treated in the Department of Orthognathic and TMJ Surgery at West China Hospital of Stomatology, Sichuan University. The patients were divided into two groups based on the primary predictor variable, which was the addition of AFM - either a single bicortical screw or a two-hole plate in addition to the single miniplate. The primary outcome variable was the condylar positional and morphological changes after mandibular advancement through BSSRO. Three-dimensional facial CT scans were obtained at three different time points (preoperatively - T0, 1 week postoperatively - T1, and 1 year postoperatively - T2) and analyzed using ITK-SNAP, 3D Slicer, and SlicerSALT software. Intergroup comparisons were conducted with an independent t-test, with a p-value of <0.05 considered significant. Correlations between the variables were estimated by Pearson correlation. The study comprised 51 patients (32 females, 19 males; mean age 25.13 ± 4.24 years), involving a total of 81 condyles (21 unilateral and 60 bilateral). There was a significant difference in long-term condylar displacement in favor of AFM along with a single miniplate (p < 0.001). The bicortical screw group recorded less condylar displacement than the two-hole plate group horizontally (0.11 mm vs 0.22 mm) and sagittally (0.03 mm vs 0.17 mm), but more vertically (0.85 mm vs 0.03 mm). Bone formation associated with AFM occurred on all condylar surfaces, compared with only three surfaces in the single miniplate group. The adjunctive method in addition to the single miniplate fixation method showed less condylar displacement and more bone apposition after mandibular advancement through BSSRO. The follow-up duration variable was the only significant determinant for volumetric changes in the condyle.

Chest CT-based automated vertebral fracture assessment using artificial intelligence and morphologic features

BACKGROUND

Spinal degeneration and vertebral compression fractures are common among the elderly that adversely affect their mobility, quality of life, lung function, and mortality. Assessment of vertebral fractures in chronic obstructive pulmonary disease (COPD) is important due to the high prevalence of osteoporosis and associated vertebral fractures in COPD.

PURPOSE

We present new automated methods for (1) segmentation and labelling of individual vertebrae in chest computed tomography (CT) images using deep learning (DL), multi-parametric freeze-and-grow (FG) algorithm, and separation of apparently fused vertebrae using intensity autocorrelation and (2) vertebral deformity fracture detection using computed vertebral height features and parametric computational modelling of an established protocol outlined for trained human experts.

METHODS

A chest CT-based automated method was developed for quantitative deformity fracture assessment following the protocol by Genant et al. The computational method was accomplished in the following steps: (1) computation of a voxel-level vertebral body likelihood map from chest CT using a trained DL network; (2) delineation and labelling of individual vertebrae on the likelihood map using an iterative multi-parametric FG algorithm; (3) separation of apparently fused vertebrae in CT using intensity autocorrelation; (4) computation of vertebral heights using contour analysis on the central anterior-posterior (AP) plane of a vertebral body; (5) assessment of vertebral fracture status using ratio functions of vertebral heights and optimized thresholds. The method was applied to inspiratory or total lung capacity (TLC) chest scans from the multi-site Genetic Epidemiology of COPD (COPDGene) (ClinicalTrials.gov: NCT00608764) study, and the performance was examined (n = 3231). One hundred and twenty scans randomly selected from this dataset were partitioned into training (n = 80) and validation (n = 40) datasets for the DL-based vertebral body classifier. Also, generalizability of the method to low dose CT imaging (n = 236) was evaluated.

RESULTS

The vertebral segmentation module achieved a Dice score of .984 as compared to manual outlining results as reference (n = 100); the segmentation performance was consistent across images with the minimum and maximum of Dice scores among images being .980 and .989, respectively. The vertebral labelling module achieved 100% accuracy (n = 100). For low dose CT, the segmentation module produced image-level minimum and maximum Dice scores of .995 and .999, respectively, as compared to standard dose CT as the reference; vertebral labelling at low dose CT was fully consistent with standard dose CT (n = 236). The fracture assessment method achieved overall accuracy, sensitivity, and specificity of 98.3%, 94.8%, and 98.5%, respectively, for 40,050 vertebrae from 3231 COPDGene participants. For generalizability experiments, fracture assessment from low dose CT was consistent with the reference standard dose CT results across all participants.

CONCLUSIONS

Our CT-based automated method for vertebral fracture assessment is accurate, and it offers a feasible alternative to manual expert reading, especially for large population-based studies, where automation is important for high efficiency. Generalizability of the method to low dose CT imaging further extends the scope of application of the method, particularly since the usage of low dose CT imaging in large population-based studies has increased to reduce cumulative radiation exposure.

Convolutional Neural Network for Fully Automated Cerebellar Volumetry in Children in Comparison to Manual Segmentation and Developmental Trajectory of Cerebellar Volumes

The purpose of this study was to develop a fully automated and reliable volumetry of the cerebellum of children during infancy and childhood using deep learning algorithms in comparison to manual segmentation. In addition, the clinical usefulness of measuring the cerebellar volume is shown. One hundred patients (0 to 16.3 years old) without infratentorial signal abnormalities on conventional MRI were retrospectively selected from our pool of pediatric MRI examinations. Based on a routinely acquired 3D T1-weighted magnetization prepared rapid gradient echo (MPRAGE) sequence, the cerebella were manually segmented using ITK-SNAP. The data set of all 100 cases was divided into four splits (four-fold cross-validation) to train the network (NN) to delineate the boundaries of the cerebellum. First, the accuracy of the newly created neural network was compared with the manual segmentation. Secondly, age-related volume changes were investigated. Our trained NN achieved an excellent Spearman correlation coefficient of 0.99, a Dice Coefficient of 95.0 ± 2.1%, and an intersection over union (IoU) of 90.6 ± 3.8%. Cerebellar volume increased continuously with age, showing an exponentially rapid growth within the first year of life. Using a convolutional neural network, it was possible to achieve reliable, fully automated cerebellar volume measurements in childhood and infancy, even when based on a relatively small cohort. In this preliminary study, age-dependent cerebellar volume changes could be acquired.

Adaptable cascaded registration for personalized maxilla completion and cleft defect volume estimation

BACKGROUND

Cone-beam computed tomography (CBCT) images provide high-resolution insights into the underlying craniofacial anomaly in patients with cleft lip and palate (CLP), requiring non-negligible annotation costs to measure the cleft defect for the guidance of the clinical secondary alveolar bone graft procedures. Considering the cumbersome volumetric image acquisition, there is a lack of paired CLP CBCTs and normal CBCTs for learning-based anatomical structure restoration models. Nowadays, the registration-based method relieves the annotation burden, though one-shot registration and the regular mask are limited to handling fine-grained shape variations and harmony between restored bony tissues and the defected maxilla.

PURPOSE

This study aimed to design and evaluate a novel method for deformable partial registration of the CLP CBCTs and normal CBCTs, enabling personalized maxilla completion and cleft defect volume prediction from CLP CBCTs.

METHODS

We proposed an adaptable deep registration framework for personalized maxilla completion and cleft defect volume prediction from CLP CBCTs. The key ingredient was a cascaded partial registration to exploit the maxillary morphology prior and attribute transfer. Cascaded registration with coarse-to-fine registration fields handled morphological variations of cleft defects and fine-grained maxillary restoration. We designed an adaptable cleft defect mask and volumetric Boolean operators for reliable voxel filling of the defected maxilla. A total of 36 clinically obtained CLP CBCTs were used to train and validate the proposed model, among which 22 CLP CBCTs were used to generate a training dataset with 440 synthetic CBCTs by B-spline deformation-based data augmentation and the remaining for testing. The proposed model was evaluated on maxilla completion and cleft defect volume prediction from clinically obtained unilateral and bilateral CLP CBCTs.

RESULTS

Extensive experiments demonstrated the effectiveness of the adaptable cleft defect mask and the cascaded partial registration on maxilla completion and cleft defect volume prediction. The proposed method achieved state-of-the-art performances with the Dice similarity coefficient of 0.90 ± $\pm$ 0.02 on the restored maxilla and 0.84 ± $\pm$ 0.04 on the estimated cleft defect, respectively. The average Hausdorff distance between the estimated cleft defect and the manually annotated ground truth was 0.30 ± $\pm$ 0.08 mm. The relative volume error of the cleft defect was0.09 ± $0.09\pm$ 0.08. The proposed model allowed for the prediction of cleft defect maps that were in line with the ground truth in the challenging unilateral and bilateral CLP CBCTs.

CONCLUSIONS

The results suggest that the proposed adaptable deep registration model enables patient-specific maxilla completion and automatic annotation of cleft defects, relieving tedious voxel-wise annotation and image acquisition burdens.

In vivo micro-computed tomography evaluation of radiopaque, polymeric device degradation in normal and inflammatory environments

Polymeric biomedical implants are an important clinical tool, but degradation remains difficult to determine post-implantation. Computed tomography (CT) could be a powerful tool for device monitoring, but polymers require incorporation of radiopaque contrast agents to be distinguishable from tissue. In addition, immune response to radiopaque devices must be characterized as it modulates device function. Radiopaque devices and films were produced by incorporating 0-20 wt% TaOx nanoparticles into polymers: polycaprolactone (PCL) and poly(lactide-co-glycolide) (PLGA). In vitro inflammatory responses of mouse bone marrow-derived macrophages to polymer matrix incorporating TaOx nanoparticles was determined by monitoring cytokine secretion. Nanoparticle addition stimulated a slight inflammatory reaction, increasing TNFα secretion, mediated by changes in polymer matrix properties. Subsequently, devices (PLGA 50:50 + 20 wt% TaOx) were implanted subcutaneously in a mouse model of chronic inflammation, that featured a sustained increase in inflammatory response local to the implant site over 12 weeks. No changes to device degradation rates or foreign body response were noted between a normal and chronically stimulated inflammatory environment. Serial CT device monitoring post-implantation provided a detailed timeline of device collapse, with no rapid, spontaneous release of nanoparticles that occluded matrix visualization. Importantly, repeat CT sessions did not ablate the immune system or alter degradation kinetics. Thus, polymer devices incorporating radiopaque nanoparticles can be used for in situ monitoring and be readily combined with other medical imaging techniques, for a dynamic view biomaterial and tissue interactions. STATEMENT OF SIGNIFICANCE: A growing number of implantable devices are in use in the clinic, exposing patients to inherent risks of implant movement, collapse, and infection. The ability to monitor implanted devices would enable faster diagnosis of failure and open the door for personalized rehabilitation therapies - both of which could vastly improve patient outcomes. Unfortunately, polymeric materials which make up most biomedical devices are not radiologically distinguishable from tissue post-implantation. The introduction of radiopaque nanoparticles into polymers allows for serial monitoring via computed tomography, without affecting device degradation. Here we demonstrate for the first time that nanoparticles do not undergo burst release from devices post-implantation and that inflammatory responses - a key determinant of device function in vivo - are also unaffected by nanoparticle addition.

The sex of organ geometry

Organs have a distinctive yet often overlooked spatial arrangement in the body1-5. We propose that there is a logic to the shape of an organ and its proximity to its neighbours. Here, by using volumetric scans of many Drosophila melanogaster flies, we develop methods to quantify three-dimensional features of organ shape, position and interindividual variability. We find that both the shapes of organs and their relative arrangement are consistent yet differ between the sexes, and identify unexpected interorgan adjacencies and left-right organ asymmetries. Focusing on the intestine, which traverses the entire body, we investigate how sex differences in three-dimensional organ geometry arise. The configuration of the adult intestine is only partially determined by physical constraints imposed by adjacent organs; its sex-specific shape is actively maintained by mechanochemical crosstalk between gut muscles and vascular-like trachea. Indeed, sex-biased expression of a muscle-derived fibroblast growth factor-like ligand renders trachea sexually dimorphic. In turn, tracheal branches hold gut loops together into a male or female shape, with physiological consequences. Interorgan geometry represents a previously unrecognized level of biological complexity which might enable or confine communication across organs and could help explain sex or species differences in organ function.

A spontaneous hyperglycaemic cynomolgus monkey presents cognitive deficits, neurological dysfunction and cataract

Chronic hyperglycaemia is a chief feature of diabetes mellitus and complicates with many systematic anomalies. Non-human primates (NHPs) are excellent for studying hyperglycaemia or diabetes and associated comorbidities, but lack behavioural observation. In the study, behavioural, brain imaging and histological analysis were performed in a case of spontaneously hyperglycaemic (HGM) Macaca fascicularis. The results were shown that the HGM monkey had persistent body weight loss, long-term hyperglycaemia, insulin resistance, dyslipidemia, but normal concentrations of insulin, C-peptide, insulin autoantibody, islet cell antibody and glutamic acid decarboxylase antibody. Importantly, an impaired working memory in a delayed response task and neurological dysfunctions were found in the HGM monkey. The tendency for atrophy in hippocampus was observed by magnetic resonance imaging. Lenticular opacification, lens fibres disruptions and vacuole formation also occurred to the HGM monkey. The data suggested that the spontaneous HGM monkey might present diabetes-like characteristics and associated neurobehavioral anomalies in this case. This study first reported cognitive deficits in a spontaneous hyperglycaemia NHPs, which might provide evidence to use macaque as a promising model for translational research in diabetes and neurological complications.

Impact of Pressure Guidewire on Model-Based FFR Prediction

INTRODUCTION

Fractional Flow Reserve (FFR) is used to characterize the functional significance of coronary artery stenoses. FFR is assessed under hyperemic conditions by invasive measurements of trans-stenotic pressure thanks to the insertion of a pressure guidewire across the coronary stenosis during catheterization. In order to overcome the potential risk related to the invasive procedure and to reduce the associated high costs, three-dimensional blood flow simulations that incorporate clinical imaging and patient-specific characteristics have been proposed.

PURPOSE

Most CCTA-derived FFR models neglect the potential influence of the guidewire on computed flow and pressure. Here we aim to quantify the impact of taking into account the presence of the guidewire in model-based FFR prediction.

METHODS

We adopt a CCTA-derived FFR model and perform simulations with and without the guidewire for 18 patients with suspected stable CAD.

RESULTS

Presented results show that the presence of the guidewire leads to a tendency to predict a lower FFR value. The FFR reduction is prominent in cases of severe stenoses, while the influence of the guidewire is less pronounced in cases of moderate stenoses.

CONCLUSION

From a clinical decision-making point of view, including of the pressure guidewire is potentially relevant only for intermediate stenosis cases.

Zero-DeepSub: Zero-shot deep subspace reconstruction for rapid multiparametric quantitative MRI using 3D-QALAS

PURPOSE

To develop and evaluate methods for (1) reconstructing 3D-quantification using an interleaved Look-Locker acquisition sequence with T2 preparation pulse (3D-QALAS) time-series images using a low-rank subspace method, which enables accurate and rapid T1 and T2 mapping, and (2) improving the fidelity of subspace QALAS by combining scan-specific deep-learning-based reconstruction and subspace modeling.

THEORY AND METHODS

A low-rank subspace method for 3D-QALAS (i.e., subspace QALAS) and zero-shot deep-learning subspace method (i.e., Zero-DeepSub) were proposed for rapid and high fidelity T1 and T2 mapping and time-resolved imaging using 3D-QALAS. Using an ISMRM/NIST system phantom, the accuracy and reproducibility of the T1 and T2 maps estimated using the proposed methods were evaluated by comparing them with reference techniques. The reconstruction performance of the proposed subspace QALAS using Zero-DeepSub was evaluated in vivo and compared with conventional QALAS at high reduction factors of up to nine-fold.

RESULTS

Phantom experiments showed that subspace QALAS had good linearity with respect to the reference methods while reducing biases and improving precision compared to conventional QALAS, especially for T2 maps. Moreover, in vivo results demonstrated that subspace QALAS had better g-factor maps and could reduce voxel blurring, noise, and artifacts compared to conventional QALAS and showed robust performance at up to nine-fold acceleration with Zero-DeepSub, which enabled whole-brain T1, T2, and PD mapping at 1 mm isotropic resolution within 2 min of scan time.

CONCLUSION

The proposed subspace QALAS along with Zero-DeepSub enabled high fidelity and rapid whole-brain multiparametric quantification and time-resolved imaging.

Molecular subtypes of breast cancer identified by dynamically enhanced MRI radiomics: the delayed phase cannot be ignored

OBJECTIVES

To compare the diagnostic performance of intratumoral and peritumoral features from different contrast phases of breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) by building radiomics models for differentiating molecular subtypes of breast cancer.

METHODS

This retrospective study included 377 patients with pathologically confirmed breast cancer. Patients were divided into training set (n = 202), validation set (n = 87) and test set (n = 88). The intratumoral volume of interest (VOI) and peritumoral VOI were delineated on primary breast cancers at three different DCE-MRI contrast phases: early, peak, and delayed. Radiomics features were extracted from each phase. After feature standardization, the training set was filtered by variance analysis, correlation analysis, and least absolute shrinkage and selection (LASSO). Using the extracted features, a logistic regression model based on each tumor subtype (Luminal A, Luminal B, HER2-enriched, triple-negative) was established. Ten models based on intratumoral or/plus peritumoral features from three different phases were developed for each differentiation.

RESULTS

Radiomics features extracted from delayed phase DCE-MRI demonstrated dominant diagnostic performance over features from other phases. However, the differences were not statistically significant. In the full fusion model for differentiating different molecular subtypes, the most frequently screened features were those from the delayed phase. According to the Shapley additive explanation (SHAP) method, the most important features were also identified from the delayed phase.

CONCLUSIONS

The intratumoral and peritumoral radiomics features from the delayed phase of DCE-MRI can provide additional information for preoperative molecular typing. The delayed phase of DCE-MRI cannot be ignored.

CRITICAL RELEVANCE STATEMENT

Radiomics features extracted and radiomics models constructed from the delayed phase of DCE-MRI played a crucial role in molecular subtype classification, although no significant difference was observed in the test cohort.

KEY POINTS

The molecular subtype of breast cancer provides a basis for setting treatment strategy and prognosis. The delayed-phase radiomics model outperformed that of early-/peak-phases, but no differently than other phases or combinations. Both intra- and peritumoral radiomics features offer valuable insights for molecular typing.

Complimentary vertebrate Wac models exhibit phenotypes relevant to DeSanto-Shinawi Syndrome

Monogenic syndromes are associated with neurodevelopmental changes that result in cognitive impairments, neurobehavioral phenotypes including autism and attention deficit hyperactivity disorder (ADHD), and seizures. Limited studies and resources are available to make meaningful headway into the underlying molecular mechanisms that result in these symptoms. One such example is DeSanto-Shinawi Syndrome (DESSH), a rare disorder caused by pathogenic variants in the WAC gene. Individuals with DESSH syndrome exhibit a recognizable craniofacial gestalt, developmental delay/intellectual disability, neurobehavioral symptoms that include autism, ADHD, behavioral difficulties and seizures. However, no thorough studies from a vertebrate model exist to understand how these changes occur. To overcome this, we developed both murine and zebrafish Wac/wac deletion mutants and studied whether their phenotypes recapitulate those described in individuals with DESSH syndrome. We show that the two Wac models exhibit craniofacial and behavioral changes, reminiscent of abnormalities found in DESSH syndrome. In addition, each model revealed impacts to GABAergic neurons and further studies showed that the mouse mutants are susceptible to seizures, changes in brain volumes that are different between sexes and relevant behaviors. Finally, we uncovered transcriptional impacts of Wac loss of function that will pave the way for future molecular studies into DESSH. These studies begin to uncover some biological underpinnings of DESSH syndrome and elucidate the biology of Wac, with advantages in each model.

Super resolution using sparse sampling at portable ultra-low field MR

Ultra-low field (ULF) magnetic resonance imaging (MRI) holds the potential to make MRI more accessible, given its cost-effectiveness, reduced power requirements, and portability. However, signal-to-noise ratio (SNR) drops with field strength, necessitating imaging with lower resolution and longer scan times. This study introduces a novel Fourier-based Super Resolution (FouSR) approach, designed to enhance the resolution of ULF MRI images with minimal increase in total scan time. FouSR combines spatial frequencies from two orthogonal ULF images of anisotropic resolution to create an isotropic T2-weighted fluid-attenuated inversion recovery (FLAIR) image. We hypothesized that FouSR could effectively recover information from under-sampled slice directions, thereby improving the delineation of multiple sclerosis (MS) lesions and other significant anatomical features. Importantly, the FouSR algorithm can be implemented on the scanner with changes to the k-space trajectory. Paired ULF (Hyperfine SWOOP, 0.064 tesla) and high field (Siemens, Skyra, 3 Tesla) FLAIR scans were collected on the same day from a phantom and a cohort of 10 participants with MS or suspected MS (6 female; mean ± SD age: 44.1 ± 4.1). ULF scans were acquired along both coronal and axial planes, featuring an in-plane resolution of 1.7 mm × 1.7 mm with a slice thickness of 5 mm. FouSR was evaluated against registered ULF coronal and axial scans, their average (ULF average) and a gold standard SR (ANTs SR). FouSR exhibited higher SNR (47.96 ± 12.6) compared to ULF coronal (36.7 ± 12.2) and higher lesion conspicuity (0.12 ± 0.06) compared to ULF axial (0.13 ± 0.07) but did not exhibit any significant differences contrast-to-noise-ratio (CNR) compared to other methods in patient scans. However, FouSR demonstrated superior image sharpness (0.025 ± 0.0040) compared to all other techniques (ULF coronal 0.021 ± 0.0037, q = 5.9, p-adj. = 0.011; ULF axial 0.018 ± 0.0026, q = 11.1, p-adj. = 0.0001; ULF average 0.019 ± 0.0034, q = 24.2, p-adj. < 0.0001) and higher lesion sharpness (-0.97 ± 0.31) when compared to the ULF average (-1.02 ± 0.37, t(543) = -10.174, p = <0.0001). Average blinded qualitative assessment by three experienced MS neurologists showed no significant difference in WML and sulci or gyri visualization between FouSR and other methods. FouSR can, in principle, be implemented on the scanner to produce clinically useful FLAIR images at higher resolution on the fly, providing a valuable tool for visualizing lesions and other anatomical structures in MS.

Radiomics for predicting MGMT status in cerebral glioblastoma: comparison of different MRI sequences

Using radiomics to predict O6-methylguanine-DNA methyltransferase promoter methylation status in patients with newly diagnosed glioblastoma and compare the performances of different MRI sequences. Preoperative MRI scans from 215 patients were included in this retrospective study. After image preprocessing and feature extraction, two kinds of machine-learning models were established and compared for their performances. One kind was established using all MRI sequences (T1-weighted image, T2-weighted image, contrast enhancement, fluid-attenuated inversion recovery, DWI_b_high, DWI_b_low and apparent diffusion coefficient), and the other kind was based on single MRI sequence as listed above. For the machine-learning model based on all sequences, a total of seven radiomic features were selected with the Maximum Relevance and Minimum Redundancy algorithm. The predictive accuracy was 0.993 and 0.750 in the training and validation sets, respectively, and the area under curves were 1.000 and 0.754 in the two sets, respectively. For the machine-learning model based on single sequence, the numbers of selected features were 8, 10, 10, 13, 9, 7 and 6 for T1-weighted image, T2-weighted image, contrast enhancement, fluid-attenuated inversion recovery, DWI_b_high, DWI_b_low and apparent diffusion coefficient, respectively, with predictive accuracies of 0.797-1.000 and 0.583-0.694 in the training and validation sets, respectively, and the area under curves of 0.874-1.000 and 0.538-0.697 in the two sets, respectively. Specifically, T1-weighted image-based model performed best, while contrast enhancement-based model performed worst in the independent validation set. The machine-learning models based on seven different single MRI sequences performed differently in predicting O6-methylguanine-DNA methyltransferase status in glioblastoma, while the machine-learning model based on the combination of all sequences performed best.

Automatic segmentation of medial temporal lobe subregions in multi-scanner, multi-modality MRI of variable quality

Background

Volumetry of subregions in the medial temporal lobe (MTL) computed from automatic segmentation in MRI can track neurodegeneration in Alzheimer's disease. However, image quality may vary in MRI. Poor quality MR images can lead to unreliable segmentation of MTL subregions. Considering that different MRI contrast mechanisms and field strengths (jointly referred to as "modalities" here) offer distinct advantages in imaging different parts of the MTL, we developed a muti-modality segmentation model using both 7 tesla (7T) and 3 tesla (3T) structural MRI to obtain robust segmentation in poor-quality images.

Method

MRI modalities including 3T T1-weighted, 3T T2-weighted, 7T T1-weighted and 7T T2-weighted (7T-T2w) of 197 participants were collected from a longitudinal aging study at the Penn Alzheimer's Disease Research Center. Among them, 7T-T2w was used as the primary modality, and all other modalities were rigidly registered to the 7T-T2w. A model derived from nnU-Net took these registered modalities as input and outputted subregion segmentation in 7T-T2w space. 7T-T2w images most of which had high quality from 25 selected training participants were manually segmented to train the multi-modality model. Modality augmentation, which randomly replaced certain modalities with Gaussian noise, was applied during training to guide the model to extract information from all modalities. To compare our proposed model with a baseline single-modality model in the full dataset with mixed high/poor image quality, we evaluated the ability of derived volume/thickness measures to discriminate Amyloid+ mild cognitive impairment (A+MCI) and Amyloid- cognitively unimpaired (A-CU) groups, as well as the stability of these measurements in longitudinal data.

Results

The multi-modality model delivered good performance regardless of 7T-T2w quality, while the single-modality model under-segmented subregions in poor-quality images. The multi-modality model generally demonstrated stronger discrimination of A+MCI versus A-CU. Intra-class correlation and Bland-Altman plots demonstrate that the multi-modality model had higher longitudinal segmentation consistency in all subregions while the single-modality model had low consistency in poor-quality images.

Conclusion

The multi-modality MRI segmentation model provides an improved biomarker for neurodegeneration in the MTL that is robust to image quality. It also provides a framework for other studies which may benefit from multimodal imaging.

Using diffusion tensor imaging to depict myocardial changes after matured pluripotent stem cell-derived cardiomyocyte transplantation

BACKGROUND

Novel treatment strategies are needed to improve the structure and function of the myocardium post-infarction. In vitro-matured pluripotent stem cell-derived cardiomyocytes (PSC-CMs) have been shown to be a promising regenerative strategy. We hypothesized that mature PSC-CMs will have anisotropic structure and improved cell alignment when compared to immature PSC-CMs using cardiovascular magnetic resonance (CMR) in a guinea pig model of cardiac injury.

METHODS

Guinea pigs (n = 16) were cryoinjured on day -10, followed by transplantation of either 108 polydimethylsiloxane (PDMS)-matured PSC-CMs (n = 6) or 108 immature tissue culture plastic (TCP)-generated PSC-CMs (n = 6) on day 0. Vehicle (sham-treated) subjects were injected with a pro-survival cocktail devoid of cells (n = 4), while healthy controls (n = 4) did not undergo cryoinjury or treatment. Animals were sacrificed on either day +14 or day +28 post-transplantation. Animals were imaged ex vivo on a 7T Bruker MRI. A 3D diffusion tensor imaging (DTI) sequence was used to quantify structure via fractional anisotropy (FA), mean diffusivity (MD), and myocyte alignment measured by the standard deviation of the transverse angle (TA).

RESULTS

MD and FA of mature PDMS grafts demonstrated anisotropy was not significantly different than the healthy control hearts (MD = 1.1 ± 0.12 × 10-3 mm2/s vs 0.93 ± 0.01 × 10-3 mm2/s, p = 0.4 and FA = 0.22 ± 0.05 vs 0.26 ± 0.001, p = 0.5). Immature TCP grafts exhibited significantly higher MD than the healthy control (1.3 ± 0.08 × 10-3 mm2/s, p < 0.05) and significantly lower FA than the control (0.12 ± 0.02, p < 0.05) but were not different from mature PDMS grafts in this small cohort. TA of healthy controls showed low variability and was not significantly different than mature PDMS grafts (p = 0.4) while immature TCP grafts were significantly different (p < 0.001). DTI parameters of mature graft tissue trended toward that of the healthy myocardium, indicating the grafted cardiomyocytes may have a similar phenotype to healthy tissue. Contrast-enhanced magnetic resonance images corresponded well to histological staining, demonstrating a non-invasive method of localizing the repopulated cardiomyocytes within the scar.

CONCLUSIONS

The DTI measures within graft tissue were indicative of anisotropic structure and showed greater myocyte organization compared to the scarred territory. These findings show that MRI is a valuable tool to assess the structural impacts of regenerative therapies.

MR-based radiomics predictive modelling of EGFR mutation and HER2 overexpression in metastatic brain adenocarcinoma: a two-centre study

OBJECTIVES

Magnetic resonance (MR)-based radiomics features of brain metastases are utilised to predict epidermal growth factor receptor (EGFR) mutation and human epidermal growth factor receptor 2 (HER2) overexpression in adenocarcinoma, with the aim to identify the most predictive MR sequence.

METHODS

A retrospective inclusion of 268 individuals with brain metastases from adenocarcinoma across two institutions was conducted. Utilising T1-weighted imaging (T1 contrast-enhanced [T1-CE]) and T2 fluid-attenuated inversion recovery (T2-FLAIR) sequences, 1,409 radiomics features were extracted. These sequences were randomly divided into training and test sets at a 7:3 ratio. The selection of relevant features was done using the least absolute shrinkage selection operator, and the training cohort's support vector classifier model was employed to generate the predictive model. The performance of the radiomics features was evaluated using a separate test set.

RESULTS

For contrast-enhanced T1-CE cohorts, the radiomics features based on 19 selected characteristics exhibited excellent discrimination. No significant differences in age, sex, and time to metastasis were observed between the groups with EGFR mutations or HER2 + and those with wild-type EGFR or HER2 (p > 0.05). Radiomics feature analysis for T1-CE revealed an area under the curve (AUC) of 0.98, classification accuracy of 0.93, sensitivity of 0.92, and specificity of 0.93 in the training cohort. In the test set, the AUC was 0.82. The 19 radiomics features for the T2-FLAIR sequence showed AUCs of 0.86 in the training set and 0.70 in the test set.

CONCLUSIONS

This study developed a T1-CE signature that could serve as a non-invasive adjunctive tool to determine the presence of EGFR mutations and HER2 + status in adenocarcinoma, aiding in the direction of treatment plans.

CLINICAL RELEVANCE STATEMENT

We propose radiomics features based on T1-CE brain MR sequences that are both evidence-based and non-invasive. These can be employed to guide clinical treatment planning in patients with brain metastases from adenocarcinoma.

Role of peritumoral tissue analysis in predicting characteristics of hepatocellular carcinoma using ultrasound-based radiomics

Predicting the biological characteristics of hepatocellular carcinoma (HCC) is essential for personalized treatment. This study explored the role of ultrasound-based radiomics of peritumoral tissues for predicting HCC features, focusing on differentiation, cytokeratin 7 (CK7) and Ki67 expression, and p53 mutation status. A cohort of 153 patients with HCC underwent ultrasound examinations and radiomics features were extracted from peritumoral tissues. Subgroups were formed based on HCC characteristics. Predictive modeling was carried out using the XGBOOST algorithm in the differentiation subgroup, logistic regression in the CK7 and Ki67 expression subgroups, and support vector machine learning in the p53 mutation status subgroups. The predictive models demonstrated robust performance, with areas under the curves of 0.815 (0.683-0.948) in the differentiation subgroup, 0.922 (0.785-1) in the CK7 subgroup, 0.762 (0.618-0.906) in the Ki67 subgroup, and 0.849 (0.667-1) in the p53 mutation status subgroup. Confusion matrices and waterfall plots highlighted the good performance of the models. Comprehensive evaluation was carried out using SHapley Additive exPlanations plots, which revealed notable contributions from wavelet filter features. This study highlights the potential of ultrasound-based radiomics, specifically the importance of peritumoral tissue analysis, for predicting HCC characteristics. The results warrant further validation of peritumoral tissue radiomics in larger, multicenter studies.

The value of the malignant subregion-based texture analysis in predicting the Ki-67 status in breast cancer

Objective

To evaluate the value of the malignant subregion-based texture analysis in predicting Ki-67 status in breast cancer.

Materials and methods

The dynamic contrast-enhanced magnetic resonance imaging data of 119 histopathologically confirmed breast cancer patients (81 patients with high Ki-67 expression status) from January 2018 to February 2023 in our hospital were retrospectively collected. According to the enhancement curve of each voxel within the tumor, three subregions were divided: washout subregion, plateau subregion, and persistent subregion. The washout subregion and the plateau subregion were merged as the malignant subregion. The texture features of the malignant subregion were extracted using Pyradiomics software for texture analysis. The differences in texture features were compared between the low and high Ki-67 expression cohorts and then the receiver operating characteristic (ROC) curve analysis to evaluate the predictive performance of texture features on Ki-67 expression. Finally, a support vector machine (SVM) classifier was constructed based on differential features to predict the expression level of Ki-67, the performance of the classifier was evaluated using ROC analysis and confirmed using 10-fold cross-validation.

Results

Through comparative analysis, 51 features exhibited significant differences between the low and high Ki-67 expression cohorts. Following feature reduction, 5 features were selected to build the SVM classifier, which achieved an area under the ROC curve (AUC) of 0.77 (0.68-0.87) for predicting the Ki-67 expression status. The accuracy, sensitivity, and specificity were 0.76, 0.80, and 0.68, respectively. The average AUC from the 10-fold cross-validation was 0.72 ± 0.14.

Conclusion

The texture features of the malignant subregion in breast cancer were potential biomarkers for predicting Ki-67 expression level in breast cancer, which might be used to precisely diagnose and guide the treatment of breast cancer.

A Primer for Utilizing Deep Learning and Abdominal MRI Imaging Features to Monitor Autosomal Dominant Polycystic Kidney Disease Progression

Abdominal imaging of autosomal dominant polycystic kidney disease (ADPKD) has historically focused on detecting complications such as cyst rupture, cyst infection, obstructing renal calculi, and pyelonephritis; discriminating complex cysts from renal cell carcinoma; and identifying sources of abdominal pain. Many imaging features of ADPKD are incompletely evaluated or not deemed to be clinically significant, and because of this, treatment options are limited. However, total kidney volume (TKV) measurement has become important for assessing the risk of disease progression (i.e., Mayo Imaging Classification) and predicting tolvaptan treatment's efficacy. Deep learning for segmenting the kidneys has improved these measurements' speed, accuracy, and reproducibility. Deep learning models can also segment other organs and tissues, extracting additional biomarkers to characterize the extent to which extrarenal manifestations complicate ADPKD. In this concept paper, we demonstrate how deep learning may be applied to measure the TKV and how it can be extended to measure additional features of this disease.

An automated hybrid approach via deep learning and radiomics focused on the midbrain and substantia nigra to detect early-stage Parkinson's disease

Objectives

The altered neuromelanin in substantia nigra pars compacta (SNpc) is a valuable biomarker in the detection of early-stage Parkinson's disease (EPD). Diagnosis via visual inspection or single radiomics based method is challenging. Thus, we proposed a novel hybrid model that integrates radiomics and deep learning methodologies to automatically detect EPD based on neuromelanin-sensitive MRI, namely short-echo-time Magnitude (setMag) reconstructed from quantitative susceptibility mapping (QSM).

Methods

In our study, we collected QSM images including 73 EPD patients and 65 healthy controls, which were stratified into training-validation and independent test sets with an 8:2 ratio. Twenty-four participants from another center were included as the external validation set. Our framework began with the detection of the brainstem utilizing YOLO-v5. Subsequently, a modified LeNet was applied to obtain deep learning features. Meanwhile, 1781 radiomics features were extracted, and 10 features were retained after filtering. Finally, the classified models based on radiomics features, deep learning features, and the hybrid of both were established through machine learning algorithms, respectively. The performance was mainly evaluated using accuracy, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). The saliency map was used to visualize the model.

Results

The hybrid feature-based support vector machine (SVM) model showed the best performance, achieving ACC of 96.3 and 95.8% in the independent test set and external validation set, respectively. The model established by hybrid features outperformed the one radiomics feature-based (NRI: 0.245, IDI: 0.112). Furthermore, the saliency map showed that the bilateral "swallow tail" sign region was significant for classification.

Conclusion

The integration of deep learning and radiomic features presents a potent strategy for the computer-aided diagnosis of EPD. This study not only validates the accuracy of our proposed model but also underscores its interpretability, evidenced by differential significance across various anatomical sites.

Rupture point is associated with divergent hemodynamics in intracranial aneurysms

Background

Understanding the risk factors leading to intracranial aneurysm (IA) rupture have still not been fully clarified. They are vital for proper medical guidance of patients harboring unruptured IAs. Clarifying the hemodynamics associated with the point of rupture could help could provide useful information about some of the risk factors. Thus far, few studies have studied this issue with often diverging conclusions.

Methods

We identified a point of rupture in patients operated for an IAs during surgery, using a combination of preoperative computed tomography (CT) and computed tomography angiography (CTA). Hemodynamic parameters were calculated both for the aneurysm sac as a whole and the point of rupture. In two cases, the results of CFD were compared with those of the experiment using particle image velocimetry (PIV).

Results

We were able to identify 6 aneurysms with a well-demarcated point of rupture. In four aneurysms, the rupture point was near the vortex with low wall shear stress (WSS) and high oscillatory shear index (OSI). In one case, the rupture point was in the flow jet with high WSS. In the last case, the rupture point was in the significant bleb and no specific hemodynamic parameters were found. The CFD results were verified in the PIV part of the study.

Conclusion

Our study shows that different hemodynamic scenarios are associated with the site of IA rupture. The numerical simulations were confirmed by laboratory models. This study further supports the hypothesis that various pathological pathways may lead to aneurysm wall damage resulting in its rupture.

Sex Matters-Insights from Testing Drug Efficacy in an Animal Model of Pancreatic Cancer

Preclinical studies rarely test the efficacy of therapies in both sexes. The field of oncology is no exception in this regard. In a model of syngeneic, orthotopic, metastasized pancreatic ductal adenocarcinoma we evaluated the impact of sex on pathological features of this disease as well as on the efficacy and possible adverse side effects of a novel, small molecule-based therapy inhibiting KRAS:SOS1, MEK1/2 and PI3K signaling in male and female C57BL/6J mice. Male mice had less tumor infiltration of CD8-positive cells, developed bigger tumors, had more lung metastasis and a lower probability of survival compared to female mice. These more severe pathological features in male animals were accompanied by higher distress at the end of the experiment. The evaluated inhibitors BI-3406, trametinib and BKM120 showed synergistic effects in vitro. This combinatorial therapy reduced tumor weight more efficiently in male animals, although the drug concentrations were similar in the tumors of both sexes. These results underline the importance of sex-specific preclinical research and at the same time provide a solid basis for future studies with the tested compounds.

Electron density dual-energy CT can improve the detection of lumbar disc herniation with higher image quality than standard and virtual non-calcium images

OBJECTIVES

To compare the diagnostic performance and image quality of dual-energy computed tomography (DECT) with electron density (ED) image reconstruction with those of DECT with standard CT (SC) and virtual non-calcium (VNCa) image reconstructions, for diagnosing lumbar disc herniation (L-HIVD).

METHODS

A total of 59 patients (354 intervertebral discs from T12/L1 to L5/S1; mean age, 60 years; 30 women and 29 men) who underwent DECT with spectral reconstruction and 3-T MRI within 2 weeks were enrolled between March 2021 and February 2022. Four radiologists independently assessed three image sets of randomized ED, SC, and VNCa images to detect L-HIVD at 8-week intervals. The coefficient of variance (CV) and the Weber contrast of the ROIs in the normal and diseased disc to cerebrospinal fluid space (NCR-normal/-diseased, respectively) were calculated to compare the image qualities of the noiseless ED and other series.

RESULTS

Overall, 129 L-HIVDs were noted on MRI. In the detection of L-HIVD, ED showed a higher AUC and sensitivity than SC and VNCa; 0.871 vs 0.807 vs 833 (p = 0.002) and 81% vs 70% vs 74% (p = 0.006 for SC), respectively. CV was much lower in all measurements of ED than those for SC and VNCa (p < 0.001). Furthermore, NCR-normal and NCR-diseased were the highest in ED (ED vs SC in NCR-normal and NCR-diseased, p =  0.001 and p = 0.004, respectively; ED vs VNCa in NCR-diseased, p = 0.044).

CONCLUSION

Compared to SC and VNCa images, DECT with ED reconstruction can enhance the AUC and sensitivity of L-HIVD detection with a lower CV and higher NCR.

CLINICAL RELEVANCE STATEMENT

To our knowledge, this is the first study to quantify the image quality of noiseless ED images. ED imaging may be helpful for detecting L-HIVD in patients who cannot undergo MRI.

KEY POINTS

ED images have diagnostic potential, but relevant quantitative analyses of image quality are limited. ED images detect disc herniation, with a better coefficient of variance and normalized contrast ratio values. ED images could detect L-HIVD when MRI is not an option.

Does Freehand, Patient-specific Instrumentation or Surgical Navigation Perform Better for Allograft Reconstruction After Tumor Resection? A Preclinical Synthetic Bone Study

BACKGROUND

Joint-sparing resection of periarticular bone tumors can be challenging because of complex geometry. Successful reconstruction of periarticular bone defects after tumor resection is often performed with structural allografts to allow for joint preservation. However, achieving a size-matched allograft to fill the defect can be challenging because allograft sizes vary, they do not always match a patient's anatomy, and cutting the allograft to perfectly fit the defect is demanding.

QUESTIONS/PURPOSES

(1) Is there a difference in mental workload among the freehand, patient-specific instrumentation, and surgical navigation approaches? (2) Is there a difference in conformance (quantitative measure of deviation from the ideal bone graft), elapsed time during reconstruction, and qualitative assessment of goodness-of-fit of the allograft reconstruction among the approaches?

METHODS

Seven surgeons used three modalities in the same order (freehand, patient-specific instrumentation, and surgical navigation) to fashion synthetic bone to reconstruct a standardized bone defect. National Aeronautics and Space Administration (NASA) mental task load index questionnaires and procedure time were captured. Cone-beam CT images of the shaped allografts were used to measure conformance (quantitative measure of deviation from the ideal bone graft) to a computer-generated ideal bone graft model. Six additional (senior) surgeons blinded to modality scored the quality of fit of the allografts into the standardized tumor defect using a 10-point Likert scale. We measured conformance using the root-mean-square metric in mm and used ANOVA for multipaired comparisons (p < 0.05 was significant).

RESULTS

There was no difference in mental NASA total task load scores among the freehand, patient-specific instrumentation, and surgical navigation techniques. We found no difference in conformance root-mean-square values (mean ± SD) between surgical navigation (2 ± 0 mm; mean values have been rounded to whole numbers) and patient-specific instrumentation (2 ± 1 mm), but both showed a small improvement compared with the freehand approach (3 ± 1 mm). For freehand versus surgical navigation, the mean difference was 1 mm (95% confidence interval [CI] 0.5 to 1.1; p = 0.01). For freehand versus patient-specific instrumentation, the mean difference was 1 mm (95% CI -0.1 to 0.9; p = 0.02). For patient-specific instrumentation versus surgical navigation, the mean difference was 0 mm (95% CI -0.5 to 0.2; p = 0.82). In evaluating the goodness of fit of the shaped grafts, we found no clinically important difference between surgical navigation (median [IQR] 7 [6 to 8]) and patient-specific instrumentation (median 6 [5 to 7.8]), although both techniques had higher scores than the freehand technique did (median 3 [2 to 4]). For freehand versus surgical navigation, the difference of medians was 4 (p < 0.001). For freehand versus patient-specific instrumentation, the difference of medians was 3 (p < 0.001). For patient-specific instrumentation versus surgical navigation, the difference of medians was 1 (p = 0.03). The mean ± procedural times for freehand was 16 ± 10 minutes, patient-specific instrumentation was 14 ± 9 minutes, and surgical navigation techniques was 24 ± 8 minutes. We found no differences in procedures times across three shaping modalities (freehand versus patient-specific instrumentation: mean difference 2 minutes [95% CI 0 to 7]; p = 0.92; freehand versus surgical navigation: mean difference 8 minutes [95% CI 0 to 20]; p = 0.23; patient-specific instrumentation versus surgical navigation: mean difference 10 minutes [95% CI 1 to 19]; p = 0.12).

CONCLUSION

Based on surgical simulation to reconstruct a standardized periarticular bone defect after tumor resection, we found a possible small advantage to surgical navigation over patient-specific instrumentation based on qualitative fit, but both techniques provided slightly better conformance of the shaped graft for fit into the standardized post-tumor resection bone defect than the freehand technique did. To determine whether these differences are clinically meaningful requires further study. The surgical navigation system presented here is a product of laboratory research development, and although not ready to be widely deployed for clinical practice, it is currently being used in a research operating room setting for patient care. This new technology is associated with a learning curve, capital costs, and potential risk. The reported preliminary results are based on a preclinical synthetic bone tumor study, which is not as realistic as actual surgical scenarios.

CLINICAL RELEVANCE

Surgical navigation systems are an emerging technology in orthopaedic and reconstruction surgery, and understanding their capabilities and limitations is paramount for clinical practice. Given our preliminary findings in a small cohort study with one scenario of standardized synthetic periarticular bone tumor defects, future investigations should include different surgical scenarios using allograft and cadaveric specimens in a more realistic surgical setting.

Correlating hippocampal and amygdala volumes with neuropathological burden in neurodegenerative diseases using 7T postmortem MRI

Numerous research groups worldwide have focused on postmortem imaging to bridge the resolution gap between clinical neuroimaging and neuropathology data. We developed a standardized protocol for brain embedding, imaging, and processing, facilitating alignment between antemortem MRI, postmortem MRI, and pathology to observe brain atrophy and structural damage progression over time. Using 7T postmortem ex vivo MRI, we explore the potential correlation of amygdala and hippocampal atrophy with neuropathological burden in both Down syndrome (DS) and Alzheimer's disease (AD) cohorts. Using 7T postmortem ex vivo MRI scans from 66 cases (12 DS and 54 AD) alongside a subset of antemortem scans (n=17), we correlated manually segmented hippocampal and amygdala volumes, adjusted for age, sex, and ApoE4 status, with pathological indicators such as Thal phase, Braak stage, limbic-predominant age-related TDP-43 encephalopathy (LATE) stage, hippocampal sclerosis (HS), and Lewy body (LB) stage. A significant correlation was observed between postmortem and antemortem volumes for the hippocampus, but a similar trend observed for the amygdala did not reach statistical significance. DS individuals exhibited notably smaller hippocampal and amygdala volumes compared to AD subjects. In DS, lower hippocampal and amygdala volumes correlated with more severe Braak stage, without significant associations with Thal phase. LATE and HS pathologies were uncommon in DS cases but trended toward smaller hippocampal volumes. In AD, lower hippocampal volume associated with dementia duration, advanced Thal phase, Braak stage, LATE stage, and HS presence, whereas reduced amygdala volume correlated mainly with severe LATE stage and HS, but not with Thal or Braak stages. No significant LB correlation was detected in either DS or AD cohorts. Hippocampal volume in AD appears influenced by both AD and LATE pathologies, while amygdala volume seems primarily influenced by LATE. In DS, smaller hippocampal volume, relative to AD, appears primarily influenced by tau pathology.

A multi-institutional meningioma MRI dataset for automated multi-sequence image segmentation

Meningiomas are the most common primary intracranial tumors and can be associated with significant morbidity and mortality. Radiologists, neurosurgeons, neuro-oncologists, and radiation oncologists rely on brain MRI for diagnosis, treatment planning, and longitudinal treatment monitoring. However, automated, objective, and quantitative tools for non-invasive assessment of meningiomas on multi-sequence MR images are not available. Here we present the BraTS Pre-operative Meningioma Dataset, as the largest multi-institutional expert annotated multilabel meningioma multi-sequence MR image dataset to date. This dataset includes 1,141 multi-sequence MR images from six sites, each with four structural MRI sequences (T2-, T2/FLAIR-, pre-contrast T1-, and post-contrast T1-weighted) accompanied by expert manually refined segmentations of three distinct meningioma sub-compartments: enhancing tumor, non-enhancing tumor, and surrounding non-enhancing T2/FLAIR hyperintensity. Basic demographic data are provided including age at time of initial imaging, sex, and CNS WHO grade. The goal of releasing this dataset is to facilitate the development of automated computational methods for meningioma segmentation and expedite their incorporation into clinical practice, ultimately targeting improvement in the care of meningioma patients.

Deep Learning-Based Detect-Then-Track Pipeline for Treatment Outcome Assessments in Immunotherapy-Treated Liver Cancer

Accurate treatment outcome assessment is crucial in clinical trials. However, due to the image-reading subjectivity, there exist discrepancies among different radiologists. The situation is common in liver cancer due to the complexity of abdominal scans and the heterogeneity of radiological imaging manifestations in liver subtypes. Therefore, we developed a deep learning-based detect-then-track pipeline that can automatically identify liver lesions from 3D CT scans then longitudinally track target lesions, thereby providing the evaluation of RECIST treatment outcomes in liver cancer. We constructed and validated the pipeline on 173 multi-national patients (344 venous-phase CT scans) consisting of a public dataset and two in-house cohorts of 28 centers. The proposed pipeline achieved a mean average precision of 0.806 and 0.726 of lesion detection on the validation and test sets. The model's diameter measurement reliability and consistency are significantly higher than that of clinicians (p = 1.6 × 10-4). The pipeline can make precise lesion tracking with accuracies of 85.7% and 90.8% then finally yield the RECIST accuracies of 82.1% and 81.4% on the validation and test sets. Our proposed pipeline can provide precise and convenient RECIST outcome assessments and has the potential to aid clinicians with more efficient therapeutic decisions.

Neuroanatomical changes observed over the course of a human pregnancy

Pregnancy is a period of profound hormonal and physiological change experienced by millions of women annually, yet the neural changes unfolding in the maternal brain throughout gestation have not been studied in humans. Leveraging precision imaging, we mapped neuroanatomical changes in an individual from preconception through two years postpartum. Pronounced decreases in gray matter volume and cortical thickness were evident across the brain, which stand in contrast to increases in white matter microstructural integrity, ventricle volume, and cerebrospinal fluid, with few regions untouched by the transition to motherhood. This dataset serves as the first comprehensive map of the human brain across gestation, providing an open-access resource for the brain imaging community to stimulate further exploration and discovery.

SAROS: A dataset for whole-body region and organ segmentation in CT imaging

The Sparsely Annotated Region and Organ Segmentation (SAROS) dataset was created using data from The Cancer Imaging Archive (TCIA) to provide a large open-access CT dataset with high-quality annotations of body landmarks. In-house segmentation models were employed to generate annotation proposals on randomly selected cases from TCIA. The dataset includes 13 semantic body region labels (abdominal/thoracic cavity, bones, brain, breast implant, mediastinum, muscle, parotid/submandibular/thyroid glands, pericardium, spinal cord, subcutaneous tissue) and six body part labels (left/right arm/leg, head, torso). Case selection was based on the DICOM series description, gender, and imaging protocol, resulting in 882 patients (438 female) for a total of 900 CTs. Manual review and correction of proposals were conducted in a continuous quality control cycle. Only every fifth axial slice was annotated, yielding 20150 annotated slices from 28 data collections. For the reproducibility on downstream tasks, five cross-validation folds and a test set were pre-defined. The SAROS dataset serves as an open-access resource for training and evaluating novel segmentation models, covering various scanner vendors and diseases.

Brain structure comparison among Parkinson disease, essential tremor, and healthy controls using 7T MRI

Both Parkinson disease (PD) and Essential tremor (ET) are movement disorders causing tremors in elderly individuals. Although PD and ET are different disease, they often present with similar initial symptoms, making their differentiation challenging with magnetic resonance imaging (MRI) techniques. This study aimed to identify structural brain differences among PD, ET, and health controls (HCs) using 7-Tesla (T) MRI. We assessed the whole-brain parcellation in gray matter volume, thickness, subcortical volume, and small regions of basal ganglia in PD (n = 18), ET (n = 15), and HCs (n = 18), who were matched for age and sex. Brain structure analysis was performed automatic segmentation through Freesurfer software. Small regions of basal ganglia were manually segmented by ITK-SNAP. Additionally, we examined the associations between clinical indicators (symptom duration, unified Parkinson diseases rating scale (UPDRS), and clinical rating scale for tremor (CRST)) and brain structure. PD showed a significant reduction in gray matter volume in the postcentral region compared to ET. ET showed a significant reduction in cerebellum volume compared to HCs. There was a negative correlation between CRST scores (B and C) and gray matter thickness in right superior frontal in ET. This study demonstrated potential of 7T MRI in differentiating brain structure differences among PD, ET, and HCs. Specific findings, such as parietal lobe atrophy in PD compared to ET and cerebellum atrophy in ET compared to HCs, the importance of advanced imaging techniques in accurately diagnosing and distinguishing between movement disorders that present with similar initial symptoms.

Preoperative Risk Stratification for Gastric Cancer: The Establishment of Dual-Energy CT-Based Radiomics Using Prospective Datasets at Two Centers

RATIONALE AND OBJECTIVES

To evaluate the performance of dual-energy CT (DECT)-based radiomics models for identifying high-risk histopathologic phenotypes-serosal invasion (pT4a), lymph node metastasis (LNM), lymphovascular invasion (LVI) and perineural invasion (PNI) in gastric cancer.

MATERIAL AND METHODS

This prospective bi-center study recruited histologically confirmed gastric adenocarcinoma patients who underwent triple-phase enhanced DECT before gastrectomy between January 2021 and July 2023. Radiomics features were extracted from polychromatic/monochromatic (40 keV, 100 keV)/iodine images at arterial/venous/delay phase, respectively. Predictive features were selected in the training dataset using logistic regression classifier, and trained models were applied to the external validation dataset. Performances of clinical models, conventional contrast enhanced CT (CECT) models and DECT models were evaluated using areas under the receiver operating characteristic curve (AUCs).

RESULTS

In total, 503 patients were recruited: 396 at training dataset (60.1 ± 10.8 years, 110 females, 286 males) and 107 at validation dataset (61.4 ± 9.5 years, 29 females, 78 males). DECT models dichotomizing pT4a, LNM, LVI, and PNI achieved AUCs of 0.891, 0.817, 0.834, and 0.889, respectively, in the validation dataset, similar with the CECT models. In the training dataset, compared to the CECT model, the DECT model provided increased performance for identifying pT4a, LNM, LVI (all P＜0.05), and similar performance for stratifying PNI (P = 0.104). The DECT models was associated with patient disease-free survival (all P＜0.05).

CONCLUSION

DECT radiomics can stratify patients preoperatively according to high-risk histopathologic phenotypes for gastric cancer and are associated with patient disease-free survival in the training dataset.

Deep learning for automatic segmentation of vestibular schwannoma: a retrospective study from multi-center routine MRI

Automatic segmentation of vestibular schwannoma (VS) from routine clinical MRI has potential to improve clinical workflow, facilitate treatment decisions, and assist patient management. Previous work demonstrated reliable automatic segmentation performance on datasets of standardized MRI images acquired for stereotactic surgery planning. However, diagnostic clinical datasets are generally more diverse and pose a larger challenge to automatic segmentation algorithms, especially when post-operative images are included. In this work, we show for the first time that automatic segmentation of VS on routine MRI datasets is also possible with high accuracy. We acquired and publicly release a curated multi-center routine clinical (MC-RC) dataset of 160 patients with a single sporadic VS. For each patient up to three longitudinal MRI exams with contrast-enhanced T1-weighted (ceT1w) (n = 124) and T2-weighted (T2w) (n = 363) images were included and the VS manually annotated. Segmentations were produced and verified in an iterative process: (1) initial segmentations by a specialized company; (2) review by one of three trained radiologists; and (3) validation by an expert team. Inter- and intra-observer reliability experiments were performed on a subset of the dataset. A state-of-the-art deep learning framework was used to train segmentation models for VS. Model performance was evaluated on a MC-RC hold-out testing set, another public VS datasets, and a partially public dataset. The generalizability and robustness of the VS deep learning segmentation models increased significantly when trained on the MC-RC dataset. Dice similarity coefficients (DSC) achieved by our model are comparable to those achieved by trained radiologists in the inter-observer experiment. On the MC-RC testing set, median DSCs were 86.2(9.5) for ceT1w, 89.4(7.0) for T2w, and 86.4(8.6) for combined ceT1w+T2w input images. On another public dataset acquired for Gamma Knife stereotactic radiosurgery our model achieved median DSCs of 95.3(2.9), 92.8(3.8), and 95.5(3.3), respectively. In contrast, models trained on the Gamma Knife dataset did not generalize well as illustrated by significant underperformance on the MC-RC routine MRI dataset, highlighting the importance of data variability in the development of robust VS segmentation models. The MC-RC dataset and all trained deep learning models were made available online.

An Automated Vertebrae Localization, Segmentation, and Osteoporotic Compression Fracture Detection Pipeline for Computed Tomographic Imaging

Osteoporosis is the most common chronic metabolic bone disease worldwide. Vertebral compression fracture (VCF) is the most common type of osteoporotic fracture. Approximately 700,000 osteoporotic VCFs are diagnosed annually in the USA alone, resulting in an annual economic burden of ~$13.8B. With an aging population, the rate of osteoporotic VCFs and their associated burdens are expected to rise. Those burdens include pain, functional impairment, and increased medical expenditure. Therefore, it is of utmost importance to develop an analytical tool to aid in the identification of VCFs. Computed Tomography (CT) imaging is commonly used to detect occult injuries. Unlike the existing VCF detection approaches based on CT, the standard clinical criteria for determining VCF relies on the shape of vertebrae, such as loss of vertebral body height. We developed a novel automated vertebrae localization, segmentation, and osteoporotic VCF detection pipeline for CT scans using state-of-the-art deep learning models to bridge this gap. To do so, we employed a publicly available dataset of spine CT scans with 325 scans annotated for segmentation, 126 of which also graded for VCF (81 with VCFs and 45 without VCFs). Our approach attained 96% sensitivity and 81% specificity in detecting VCF at the vertebral-level, and 100% accuracy at the subject-level, outperforming deep learning counterparts tested for VCF detection without segmentation. Crucially, we showed that adding predicted vertebrae segments as inputs significantly improved VCF detection at both vertebral and subject levels by up to 14% Sensitivity and 20% Specificity (p-value = 0.028).

Two for One-Combined Morphologic and Quantitative Knee Joint MRI Using a Versatile Turbo Spin-Echo Platform

Quantitative MRI techniques such as T2 and T1ρ mapping are beneficial in evaluating knee joint pathologies; however, long acquisition times limit their clinical adoption. MIXTURE (Multi-Interleaved X-prepared Turbo Spin-Echo with IntUitive RElaxometry) provides a versatile turbo spin-echo (TSE) platform for simultaneous morphologic and quantitative joint imaging. Two MIXTURE sequences were designed along clinical requirements: "MIX1", combining proton density (PD)-weighted fat-saturated (FS) images and T2 mapping (acquisition time: 4:59 min), and "MIX2", combining T1-weighted images and T1ρ mapping (6:38 min). MIXTURE sequences and their reference 2D and 3D TSE counterparts were acquired from ten human cadaveric knee joints at 3.0 T. Contrast, contrast-to-noise ratios, and coefficients of variation were comparatively evaluated using parametric tests. Clinical radiologists (n = 3) assessed diagnostic quality as a function of sequence and anatomic structure using five-point Likert scales and ordinal regression, with a significance level of α = 0.01. MIX1 and MIX2 had at least equal diagnostic quality compared to reference sequences of the same image weighting. Contrast, contrast-to-noise ratios, and coefficients of variation were largely similar for the PD-weighted FS and T1-weighted images. In clinically feasible scan times, MIXTURE sequences yield morphologic, TSE-based images of diagnostic quality and quantitative parameter maps with additional insights on soft tissue composition and ultrastructure.

Versican controlled by Lmx1b regulates hyaluronate density and hydration for semicircular canal morphogenesis

During inner ear semicircular canal morphogenesis in zebrafish, patterned canal-genesis zones express genes for extracellular matrix component synthesis. These include hyaluronan and the hyaluronan-binding chondroitin sulfate proteoglycan Versican, which are abundant in the matrices of many developing organs. Charged hyaluronate polymers play a key role in canal morphogenesis through osmotic swelling. However, the developmental factor(s) that control the synthesis of the matrix components and regulation of hyaluronate density and swelling are unknown. Here, we identify the transcription factor, Lmx1b, as a positive transcriptional regulator of hyaluronan, Versican, and chondroitin synthesis genes crucial for canal morphogenesis. We show that Versican regulates hyaluronan density through its protein core, whereas the charged chondroitin side chains contribute to the osmotic swelling of hyaluronate. Versican-tuned properties of hyaluronate matrices may be a broadly used mechanism in morphogenesis with important implications for understanding diseases where these matrices are impaired, and for hydrogel engineering for tissue regeneration.