Spatial normalization and quantification approaches of PET imaging for neurological disorders

Deep Learning-Based Precontrast CT Parcellation for MRI-Free Brain Amyloid PET Quantification

PURPOSE

This study aimed to develop a deep learning (DL) model for brain region parcellation using CT data from PET/CT scans to enable accurate amyloid quantification in 18F-FBB PET/CT without relying on high-resolution MRI.

PATIENTS AND METHODS

A retrospective dataset of PET/CT and T1-weighted MRI pairs from 226 individuals (157 with mild cognitive impairment or dementia and 69 healthy controls) was used. The dataset was split into training/validation (60%) and test (40%) sets. Utilizing auto-generated segmentation labels, 3 UNets were independently trained for multiplanar brain parcellation on CT and subsequently ensembled. Amyloid load was measured across 46 volumes of interest (VOIs), derived from the Desikan-Killiany-Tourville atlas. Dice similarity coefficient between the proposed CT-based DL model and MRI-based (FreeSurfer) method was calculated, with SUVR comparison using linear regression analysis and intraclass correlation coefficient. Global SUVRs were also compared within groups with clinical dementia ratings (CDRs) of 0, 0.5, and 1.

RESULTS

The DL-based CT parcellation achieved mean Dice similarity coefficients of 0.80 for all 46 VOIs, 0.72 for 16 cortical and limbic VOIs, and 0.83 for 30 subcortical VOIs. For regional and global SUVR comparisons, the linear regression yielded a slope, y-intercept, and R2 of 1 ± 0.027, 0 ± 0.040, and ≧0.976, respectively (P < 0.001), and the intraclass correlation coefficient was ≧0.988 (P < 0.001). For global SUVRs in each CDR group, these values were 1 ± 0.020, 0 ± 0.026, ≧0.993, and ≧0.996, respectively (P < 0.001). Both MRI-based and CT-based global SUVR showed a consistent increase as the CDR score increased.

CONCLUSIONS

The DL-based CT parcellation agrees strongly with MRI-based methods for amyloid PET quantification.

Investigating resting-state functional connectivity changes within procedural memory network across neuropsychiatric disorders using fMRI

BACKGROUND

Cognitive networks impairments are common in neuropsychiatric disorders like Attention Deficit Hyperactivity Disorder (ADHD), bipolar disorder (BD), and schizophrenia (SZ). While previous research has focused on specific brain regions, the role of the procedural memory as a type of long-term memory to examine cognitive networks impairments in these disorders remains unclear. This study investigates alterations in resting-state functional connectivity (rs-FC) within the procedural memory network to explore brain function associated with cognitive networks in patients with these disorders.

METHODS

This study analyzed resting-state functional magnetic resonance imaging (rs-fMRI) data from 40 individuals with ADHD, 49 with BD, 50 with SZ, and 50 healthy controls (HCs). A procedural memory network was defined based on the selection of 34 regions of interest (ROIs) associated with the network in the Harvard-Oxford Cortical Structural Atlas (default atlas). Multivariate region of interest to region of interest connectivity (mRRC) was used to analyze the rs-FC between the defined network regions. Significant differences in rs-FC between patients and HCs were identified (P < 0.001).

RESULTS

ADHD patients showed increased Cereb45 l - Cereb3 r rs-FC (p = 0.000067) and decreased Cereb1 l - Cereb6 l rs-FC (p = 0.00092). BD patients exhibited increased rs-FC between multiple regions, including Claustrum r - Caudate r (p = 0.00058), subthalamic nucleus r - Pallidum l (p = 0.00060), substantia nigra l - Cereb2 l (p = 0.00082), Cereb10 r - SMA r (p = 0.00086), and Cereb9 r - SMA l (p = 0.00093) as well as decreased rs-FC in subthalamic nucleus r - Cereb6 l (p = 0.00013) and Cereb9 r - Cereb9 l (p = 0.00033). SZ patients indicated increased Caudate r- putamen l rs-FC (p = 0.00057) and decreased rs-FC in subthalamic nucleus r - Cereb6 l (p = 0.000063), and Cereb1 r - subthalamic nucleus r (p = 0.00063).

CONCLUSIONS

This study found significant alterations in rs-FC within the procedural memory network in patients with ADHD, BD, and SZ compared to HCs. These findings suggest that disrupted rs-FC within this network may related to cognitive networks impairments observed in these disorders.

CLINICAL TRIAL NUMBER

Not applicable.

Machine Learning Driven by Magnetic Resonance Imaging for the Classification of Alzheimer Disease Progression: Systematic Review and Meta-Analysis

BACKGROUND

To diagnose Alzheimer disease (AD), individuals are classified according to the severity of their cognitive impairment. There are currently no specific causes or conditions for this disease.

OBJECTIVE

The purpose of this systematic review and meta-analysis was to assess AD prevalence across different stages using machine learning (ML) approaches comprehensively.

METHODS

The selection of papers was conducted in 3 phases, as per PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) 2020 guidelines: identification, screening, and final inclusion. The final analysis included 24 papers that met the criteria. The selection of ML approaches for AD diagnosis was rigorously based on their relevance to the investigation. The prevalence of patients with AD at 2, 3, 4, and 6 stages was illustrated through the use of forest plots.

RESULTS

The prevalence rate for both cognitively normal (CN) and AD across 6 studies was 49.28% (95% CI 46.12%-52.45%; P=.32). The prevalence estimate for the 3 stages of cognitive impairment (CN, mild cognitive impairment, and AD) is 29.75% (95% CI 25.11%-34.84%, P<.001). Among 5 studies with 14,839 participants, the analysis of 4 stages (nondemented, moderately demented, mildly demented, and AD) found an overall prevalence of 13.13% (95% CI 3.75%-36.66%; P<.001). In addition, 4 studies involving 3819 participants estimated the prevalence of 6 stages (CN, significant memory concern, early mild cognitive impairment, mild cognitive impairment, late mild cognitive impairment, and AD), yielding a prevalence of 23.75% (95% CI 12.22%-41.12%; P<.001).

CONCLUSIONS

The significant heterogeneity observed across studies reveals that demographic and setting characteristics are responsible for the impact on AD prevalence estimates. This study shows how ML approaches can be used to describe AD prevalence across different stages, which provides valuable insights for future research.

CT-guided spatial normalization of nuclear hybrid imaging adapted to enlarged ventricles: Impact on striatal uptake quantification

INTRODUCTION

Spatial normalization is a prerequisite step for the quantitative analysis of SPECT or PET brain images using volume-of-interest (VOI) template or voxel-based analysis. MRI-guided spatial normalization is the gold standard, but the wide use of PET/CT or SPECT/CT in routine clinical practice makes CT-guided spatial normalization a necessary alternative. Ventricular enlargement is observed with aging, and it hampers the spatial normalization of the lateral ventricles and striatal regions, limiting their analysis. The aim of the present study was to propose a robust spatial normalization method based on CT scans that takes into account features of the aging brain to reduce bias in the CT-guided striatal analysis of SPECT images.

METHODS

We propose an enhanced CT-guided spatial normalization pipeline based on SPM12. Performance of the proposed pipeline was assessed on visually normal [123I]-FP-CIT SPECT/CT images. SPM12 default CT-guided spatial normalization was used as reference method. The metrics assessed were the overlap between the spatially normalized lateral ventricles and caudate/putamen VOIs, and the computation of caudate and putamen specific binding ratios (SBR).

RESULTS

In total 231 subjects (mean age ± SD = 61.9 ± 15.5 years) were included in the statistical analysis. The mean overlap between the spatially normalized lateral ventricles of subjects and the caudate VOI and the mean SBR of caudate were respectively 38.40 % (± SD = 19.48 %) of the VOI and 1.77 (± 0.79) when performing SPM12 default spatial normalization. The mean overlap decreased to 9.13 % (± SD = 1.41 %, P < 0.001) of the VOI and the SBR of caudate increased to 2.38 (± 0.51, P < 0.0001) when performing the proposed pipeline. Spatially normalized lateral ventricles did not overlap with putamen VOI using either method. The mean putamen SBR value derived from the proposed spatial normalization (2.75 ± 0.54) was not significantly different from that derived from the default SPM12 spatial normalization (2.83 ± 0.52, P > 0.05).

CONCLUSION

The automatic CT-guided spatial normalization used herein led to a less biased spatial normalization of SPECT images, hence an improved semi-quantitative analysis. The proposed pipeline could be implemented in clinical routine to perform a more robust SBR computation using hybrid imaging.

Automated extraction of the arterial input function from brain images for parametric PET studies

BACKGROUND

Accurate measurement of the arterial input function (AIF) is crucial for parametric PET studies, but the AIF is commonly derived from invasive arterial blood sampling. It is possible to use an image-derived input function (IDIF) obtained by imaging a large blood pool, but IDIF measurement in PET brain studies performed on standard field of view scanners is challenging due to lack of a large blood pool in the field-of-view. Here we describe a novel automated approach to estimate the AIF from brain images.

RESULTS

Total body 18F-FDG PET data from 12 subjects were split into a model adjustment group (n = 6) and a validation group (n = 6). We developed an AIF estimation framework using wavelet-based methods and unsupervised machine learning to distinguish arterial and venous activity curves, compared to the IDIF from the descending aorta. All of the automatically extracted AIFs in the validation group had similar shape to the IDIF derived from the descending aorta IDIF. The average area under the curve error and normalised root mean square error across validation data were - 1.59 ± 2.93% and 0.17 ± 0.07.

CONCLUSIONS

Our automated AIF framework accurately estimates the AIF from brain images. It reduces operator-dependence, and could facilitate the clinical adoption of parametric PET.

Deep-learning predicted PET can be subtracted from the true clinical fluorodeoxyglucose PET co-registered to MRI to identify the epileptogenic zone in focal epilepsy

OBJECTIVE

Normal interictal [18 F]FDG-PET can be predicted from the corresponding T1w MRI with Generative Adversarial Networks (GANs). A technique we call SIPCOM (Subtraction Interictal PET Co-registered to MRI) can then be used to compare epilepsy patients' predicted and clinical PET. We assessed the ability of SIPCOM to identify the Resection Zone (RZ) in patients with drug-resistant epilepsy (DRE) with reference to visual and statistical parametric mapping (SPM) analysis.

METHODS

Patients with complete presurgical work-up and subsequent SEEG and cortectomy were included. RZ localisation, the reference region, was assigned to one of eighteen anatomical brain regions. SIPCOM was implemented using healthy controls to train a GAN. To compare, the clinical PET coregistered to MRI was visually assessed by two trained readers, and a standard SPM analysis was performed.

RESULTS

Twenty patients aged 17-50 (32 ± 7.8) years were included, 14 (70%) with temporal lobe epilepsy (TLE). Eight (40%) were MRI-negative. After surgery, 14 patients (70%) had a good outcome (Engel I-II). RZ localisation rate was 60% with SIPCOM vs 35% using SPM (P = 0.015) and vs 85% using visual analysis (P = 0.54). Results were similar for Engel I-II patients, the RZ localisation rate was 64% with SIPCOM vs 36% with SPM. With SIPCOM localisation was correct in 67% in MRI-positive vs 50% in MRI-negative patients, and 64% in TLE vs 43% in extra-TLE. The average number of false-positive clusters was 2.2 ± 1.3 using SIPCOM vs 2.3 ± 3.1 using SPM. All RZs localized with SPM were correctly localized with SIPCOM. In one case, PET and MRI were visually reported as negative, but both SIPCOM and SPM localized the RZ.

SIGNIFICANCE

SIPCOM performed better than the reference computer-assisted method (SPM) for RZ detection in a group of operated DRE patients. SIPCOM's impact on epilepsy management needs to be prospectively validated.

Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier

While 18F-florzolotau tau PET is an emerging biomarker for progressive supranuclear palsy (PSP), its interpretation has been hindered by a lack of consensus on visual reading and potential biases in conventional semi-quantitative analysis. As clinical manifestations and regions of elevated 18F-florzolotau binding are highly overlapping in PSP and the Parkinsonian type of multiple system atrophy (MSA-P), developing a reliable discriminative classifier for 18F-florzolotau PET is urgently needed. Herein, we developed a normalization-free deep-learning (NFDL) model for 18F-florzolotau PET, which achieved significantly higher accuracy for both PSP and MSA-P compared to semi-quantitative classifiers. Regions driving the NFDL classifier's decision were consistent with disease-specific topographies. NFDL-guided radiomic features correlated with clinical severity of PSP. This suggests that the NFDL model has the potential for early and accurate differentiation of atypical parkinsonism and that it can be applied in various scenarios due to not requiring subjective interpretation, MR-dependent, and reference-based preprocessing.

Textural properties of microglial activation in Alzheimer's disease as measured by (R)-[11C]PK11195 PET

Alzheimer's disease is the most common form of dementia worldwide, accounting for 60-70% of diagnosed cases. According to the current understanding of molecular pathogenesis, the main hallmarks of this disease are the abnormal accumulation of amyloid plaques and neurofibrillary tangles. Therefore, biomarkers reflecting these underlying biological mechanisms are recognized as valid tools for an early diagnosis of Alzheimer's disease. Inflammatory mechanisms, such as microglial activation, are known to be involved in Alzheimer's disease onset and progression. This activated state of the microglia is associated with increased expression of the translocator protein 18 kDa. On that account, PET tracers capable of measuring this signature, such as (R)-[¹¹C]PK11195, might be instrumental in assessing the state and evolution of Alzheimer's disease. This study aims to investigate the potential of Gray Level Co-occurrence Matrix-based textural parameters as an alternative to conventional quantification using kinetic models in (R)-[¹¹C]PK11195 PET images. To achieve this goal, kinetic and textural parameters were computed on (R)-[¹¹C]PK11195 PET images of 19 patients with an early diagnosis of Alzheimer's disease and 21 healthy controls and submitted separately to classification using a linear support vector machine. The classifier built using the textural parameters showed no inferior performance compared to the classical kinetic approach, yielding a slightly larger classification accuracy (accuracy of 0.7000, sensitivity of 0.6957, specificity of 0.7059 and balanced accuracy of 0.6967). In conclusion, our results support the notion that textural parameters may be an alternative to conventional quantification using kinetic models in (R)-[¹¹C]PK11195 PET images. The proposed quantification method makes it possible to use simpler scanning procedures, which increase patient comfort and convenience. We further speculate that textural parameters may also provide an alternative to kinetic analysis in (R)-[¹¹C]PK11195 PET neuroimaging studies involving other neurodegenerative disorders. Finally, we recognize that the potential role of this tracer is not in diagnosis but rather in the assessment and progression of the diffuse and dynamic distribution of inflammatory cell density in this disorder as a promising therapeutic target.

Feasibility of 18F-florzolotau quantification in patients with Alzheimer's disease based on an MRI-free tau PET template

OBJECTIVES

Quantification of tau accumulation using positron emission tomography (PET) is critical for the diagnosis of Alzheimer's disease (AD). This study aimed to evaluate the feasibility of ¹⁸F-florzolotau quantification in patients with AD using a magnetic resonance imaging (MRI)-free tau PET template, since individual high-resolution MRI is costly and not always available in practice.

METHODS

¹⁸F-florzolotau PET and MRI scans were obtained in a discovery cohort including (1) patients within the AD continuum (n = 87), (2) cognitively impaired patients with non-AD (n = 32), and (3) cognitively unimpaired subjects (n = 26). The validation cohort comprised 24 patients with AD. Following MRI-dependent spatial normalization (standard approach) in randomly selected subjects (n = 40) to cover the entire spectrum of cognitive function, selected PET images were averaged to create the ¹⁸F-florzolotau-specific template. Standardized uptake value ratios (SUVRs) were calculated in five predefined regions of interest (ROIs). MRI-free and MRI-dependent methods were compared in terms of continuous and dichotomous agreement, diagnostic performances, and associations with specific cognitive domains.

RESULTS

MRI-free SUVRs had a high continuous and dichotomous agreement with MRI-dependent measures for all ROIs (intraclass correlation coefficient ≥ 0.980; agreement ≥ 94.5%). Similar findings were observed for AD-related effect sizes, diagnostic performances with respect to categorization across the cognitive spectrum, and associations with cognitive domains. The robustness of the MRI-free approach was confirmed in the validation cohort.

CONCLUSIONS

The use of an ¹⁸F-florzolotau-specific template is a valid alternative to MRI-dependent spatial normalization, improving the clinical generalizability of this second-generation tau tracer.

KEY POINTS

• Regional ¹⁸F-florzolotau SUVRs reflecting tau accumulation in the living brains are reliable biomarkers for the diagnosis, differential diagnosis, and assessment of disease severity in patients with AD. • The ¹⁸F-florzolotau-specific template is a valid alternative to MRI-dependent spatial normalization, improving the clinical generalizability of this second-generation tau tracer.

A clinical trial of super-stable homogeneous lipiodol-nanoICG formulation-guided precise fluorescent laparoscopic hepatocellular carcinoma resection

BACKGROUND

Applying traditional fluorescence navigation technologies in hepatocellular carcinoma is severely restricted by high false-positive rates, variable tumor differentiation, and unstable fluorescence performance.

RESULTS

In this study, a green, economical and safe nanomedicine formulation technology was developed to construct carrier-free indocyanine green nanoparticles (nanoICG) with a small uniform size and better fluorescent properties without any molecular structure changes compared to the ICG molecule. Subsequently, nanoICG dispersed into lipiodol via a super-stable homogeneous intermixed formulation technology (SHIFT&nanoICG) for transhepatic arterial embolization combined with fluorescent laparoscopic hepatectomy to eliminate the existing shortcomings. A 52-year-old liver cancer patient was recruited for the clinical trial of SHIFT&nanoICG. We demonstrate that SHIFT&nanoICG could accurately identify and mark the lesion with excellent stability, embolism, optical imaging performance, and higher tumor-to-normal tissue ratio, especially in the detection of the microsatellite lesions (0.4 × 0.3 cm), which could not be detected by preoperative imaging, to realize a complete resection of hepatocellular carcinoma under fluorescence laparoscopy in a shorter period (within 2 h) and with less intraoperative blood loss (50 mL).

CONCLUSIONS

This simple and effective strategy integrates the diagnosis and treatment of hepatocellular carcinoma, and thus, it has great potential in various clinical applications.