Deep transformer-based personalized dosimetry from SPECT/CT images: a hybrid approach for [177Lu]Lu-DOTATATE radiopharmaceutical therapy

PURPOSE

Accurate dosimetry is critical for ensuring the safety and efficacy of radiopharmaceutical therapies. In current clinical dosimetry practice, MIRD formalisms are widely employed. However, with the rapid advancement of deep learning (DL) algorithms, there has been an increasing interest in leveraging the calculation speed and automation capabilities for different tasks. We aimed to develop a hybrid transformer-based deep learning (DL) model that incorporates a multiple voxel S-value (MSV) approach for voxel-level dosimetry in [177Lu]Lu-DOTATATE therapy. The goal was to enhance the performance of the model to achieve accuracy levels closely aligned with Monte Carlo (MC) simulations, considered as the standard of reference. We extended our analysis to include MIRD formalisms (SSV and MSV), thereby conducting a comprehensive dosimetry study.

METHODS

We used a dataset consisting of 22 patients undergoing up to 4 cycles of [177Lu]Lu-DOTATATE therapy. MC simulations were used to generate reference absorbed dose maps. In addition, MIRD formalism approaches, namely, single S-value (SSV) and MSV techniques, were performed. A UNEt TRansformer (UNETR) DL architecture was trained using five-fold cross-validation to generate MC-based dose maps. Co-registered CT images were fed into the network as input, whereas the difference between MC and MSV (MC-MSV) was set as output. DL results are then integrated to MSV to revive the MC dose maps. Finally, the dose maps generated by MSV, SSV, and DL were quantitatively compared to the MC reference at both voxel level and organ level (organs at risk and lesions).

RESULTS

The DL approach showed slightly better performance (voxel relative absolute error (RAE) = 5.28 ± 1.32) compared to MSV (voxel RAE = 5.54 ± 1.4) and outperformed SSV (voxel RAE = 7.8 ± 3.02). Gamma analysis pass rates were 99.0 ± 1.2%, 98.8 ± 1.3%, and 98.7 ± 1.52% for DL, MSV, and SSV approaches, respectively. The computational time for MC was the highest (~2 days for a single-bed SPECT study) compared to MSV, SSV, and DL, whereas the DL-based approach outperformed the other approaches in terms of time efficiency (3 s for a single-bed SPECT). Organ-wise analysis showed absolute percent errors of 1.44 ± 3.05%, 1.18 ± 2.65%, and 1.15 ± 2.5% for SSV, MSV, and DL approaches, respectively, in lesion-absorbed doses.

CONCLUSION

A hybrid transformer-based deep learning model was developed for fast and accurate dose map generation, outperforming the MIRD approaches, specifically in heterogenous regions. The model achieved accuracy close to MC gold standard and has potential for clinical implementation for use on large-scale datasets.

Development of novel low-cost readout electronics for large field-of-view gamma camera detectors

PURPOSE

Large scintillation crystals-based gamma cameras play a crucial role in nuclear medicine imaging. In this study, a large field-of-view (FOV) gamma detector consisting of 48 square PMTs developed using a new readout electronics, reducing 48 (6 × 8) analog signals to 14 (6 + 8) analog sums of each row and column, with reduced complexity and cost while preserving image quality.

METHODS

All 14 analog signals were converted to digital signals using AD9257 high-speed analog to digital (ADC) converters driven by the SPARTAN-6 family of field-programmable gate arrays (FPGA) in order to calculate the signal integrals. The positioning algorithm was based on the digital correlated signal enhancement (CSE) algorithm implemented in the acquisition software. The performance characteristics of the developed gamma camera were measured using the NEMA NU 1-2018 standards.

RESULTS

The measured energy resolution of the developed detector was 8.7 % at 140 keV, with an intrinsic spatial resolution of 3.9 mm. The uniformity was within 0.6 %, while the linearity was within 0.1 %.

CONCLUSION

The performance evaluation demonstrated that the developed detector has suitable specifications for high-end nuclear medicine imaging.

The effect of harmonization on the variability of PET radiomic features extracted using various segmentation methods

PURPOSE

This study aimed to examine the robustness of positron emission tomography (PET) radiomic features extracted via different segmentation methods before and after ComBat harmonization in patients with non-small cell lung cancer (NSCLC).

METHODS

We included 120 patients (positive recurrence = 46 and negative recurrence = 74) referred for PET scanning as a routine part of their care. All patients had a biopsy-proven NSCLC. Nine segmentation methods were applied to each image, including manual delineation, K-means (KM), watershed, fuzzy-C-mean, region-growing, local active contour (LAC), and iterative thresholding (IT) with 40, 45, and 50% thresholds. Diverse image discretizations, both without a filter and with different wavelet decompositions, were applied to PET images. Overall, 6741 radiomic features were extracted from each image (749 radiomic features from each segmented area). Non-parametric empirical Bayes (NPEB) ComBat harmonization was used to harmonize the features. Linear Support Vector Classifier (LinearSVC) with L1 regularization For feature selection and Support Vector Machine classifier (SVM) with fivefold nested cross-validation was performed using StratifiedKFold with 'n_splits' set to 5 to predict recurrence in NSCLC patients and assess the impact of ComBat harmonization on the outcome.

RESULTS

From 749 extracted radiomic features, 206 (27%) and 389 (51%) features showed excellent reliability (ICC ≥ 0.90) against segmentation method variation before and after NPEB ComBat harmonization, respectively. Among all, 39 features demonstrated poor reliability, which declined to 10 after ComBat harmonization. The 64 fixed bin widths (without any filter) and wavelets (LLL)-based radiomic features set achieved the best performance in terms of robustness against diverse segmentation techniques before and after ComBat harmonization. The first-order and GLRLM and also first-order and NGTDM feature families showed the largest number of robust features before and after ComBat harmonization, respectively. In terms of predicting recurrence in NSCLC, our findings indicate that using ComBat harmonization can significantly enhance machine learning outcomes, particularly improving the accuracy of watershed segmentation, which initially had fewer reliable features than manual contouring. Following the application of ComBat harmonization, the majority of cases saw substantial increase in sensitivity and specificity.

CONCLUSION

Radiomic features are vulnerable to different segmentation methods. ComBat harmonization might be considered a solution to overcome the poor reliability of radiomic features.

Semi-supervised learning towards automated segmentation of PET images with limited annotations: application to lymphoma patients

Manual segmentation poses a time-consuming challenge for disease quantification, therapy evaluation, treatment planning, and outcome prediction. Convolutional neural networks (CNNs) hold promise in accurately identifying tumor locations and boundaries in PET scans. However, a major hurdle is the extensive amount of supervised and annotated data necessary for training. To overcome this limitation, this study explores semi-supervised approaches utilizing unlabeled data, specifically focusing on PET images of diffuse large B-cell lymphoma (DLBCL) and primary mediastinal large B-cell lymphoma (PMBCL) obtained from two centers. We considered 2-[18F]FDG PET images of 292 patients PMBCL (n = 104) and DLBCL (n = 188) (n = 232 for training and validation, and n = 60 for external testing). We harnessed classical wisdom embedded in traditional segmentation methods, such as the fuzzy clustering loss function (FCM), to tailor the training strategy for a 3D U-Net model, incorporating both supervised and unsupervised learning approaches. Various supervision levels were explored, including fully supervised methods with labeled FCM and unified focal/Dice loss, unsupervised methods with robust FCM (RFCM) and Mumford-Shah (MS) loss, and semi-supervised methods combining FCM with supervised Dice loss (MS + Dice) or labeled FCM (RFCM + FCM). The unified loss function yielded higher Dice scores (0.73 ± 0.11; 95% CI 0.67-0.8) than Dice loss (p value < 0.01). Among the semi-supervised approaches, RFCM + αFCM (α = 0.3) showed the best performance, with Dice score of 0.68 ± 0.10 (95% CI 0.45-0.77), outperforming MS + αDice for any supervision level (any α) (p < 0.01). Another semi-supervised approach with MS + αDice (α = 0.2) achieved Dice score of 0.59 ± 0.09 (95% CI 0.44-0.76) surpassing other supervision levels (p < 0.01). Given the time-consuming nature of manual delineations and the inconsistencies they may introduce, semi-supervised approaches hold promise for automating medical imaging segmentation workflows.

An Update on the Role of mpMRI and 68Ga-PSMA PET Imaging in Primary and Recurrent Prostate Cancer

The objective of this work was to review comparisons of the efficacy of 68Ga-PSMA-11 (prostate-specific membrane antigen) PET/CT and multiparametric magnetic resonance imaging (mpMRI) in the detection of prostate cancer among patients undergoing initial staging prior to radical prostatectomy or experiencing recurrent prostate cancer, based on histopathological data. A comprehensive search was conducted in PubMed and Web of Science, and relevant articles were analyzed with various parameters, including year of publication, study design, patient count, age, PSA (prostate-specific antigen) value, Gleason score, standardized uptake value (SUVmax), detection rate, treatment history, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and PI-RADS (prostate imaging reporting and data system) scores. Only studies directly comparing PSMA-PET and mpMRI were considered, while those examining combined accuracy or focusing on either modality alone were excluded. In total, 24 studies comprising 1717 patients were analyzed, with the most common indication for screening being staging, followed by relapse. The findings indicated that 68Ga-PSMA-PET/CT effectively diagnosed prostate cancer in patients with suspected or confirmed disease, and both methods exhibited comparable efficacy in identifying lesion-specific information. However, notable heterogeneity was observed, highlighting the necessity for standardization of imaging and histopathology systems to mitigate inter-study variability. Future research should prioritize evaluating the combined diagnostic performance of both modalities to enhance sensitivity and reduce unnecessary biopsies. Overall, the utilization of PSMA-PET and mpMRI in combination holds substantial potential for significantly advancing the diagnosis and management of prostate cancer.

Hierarchical approach for pulmonary-nodule identification from CT images using YOLO model and a 3D neural network classifier

This study aimed to assist doctors in detecting early-stage lung cancer. To achieve this, a hierarchical system that can detect nodules in the lungs using computed tomography (CT) images was developed. In the initial phase, a preexisting model (YOLOv5s) was used to detect lung nodules. A 0.3 confidence threshold was established for identifying nodules in this phase to enhance the model's sensitivity. The primary objective of the hierarchical model was to locate and categorize all lung nodules while minimizing the false-negative rate. Following the analysis of the results from the first phase, a novel 3D convolutional neural network (CNN) classifier was developed to examine and categorize the potential nodules detected by the YOLOv5s model. The objective was to create a detection framework characterized by an extremely low false positive rate and high accuracy. The Lung Nodule Analysis 2016 (LUNA 16) dataset was used to evaluate the effectiveness of this framework. This dataset comprises 888 CT scans that include the positions of 1186 nodules and 400,000 non-nodular regions in the lungs. The YOLOv5s technique yielded numerous incorrect detections owing to its low confidence level. Nevertheless, the addition of a 3D classification system significantly enhanced the precision of nodule identification. By integrating the outcomes of the YOLOv5s approach using a 30% confidence limit and the 3D CNN classification model, the overall system achieved 98.4% nodule detection accuracy and an area under the curve of 98.9%. Despite producing some false negatives and false positives, the suggested method for identifying lung nodules from CT scans is promising as a valuable aid in decision-making for nodule detection.

Differential privacy preserved federated learning for prognostic modeling in COVID-19 patients using large multi-institutional chest CT dataset

BACKGROUND

Notwithstanding the encouraging results of previous studies reporting on the efficiency of deep learning (DL) in COVID-19 prognostication, clinical adoption of the developed methodology still needs to be improved. To overcome this limitation, we set out to predict the prognosis of a large multi-institutional cohort of patients with COVID-19 using a DL-based model.

PURPOSE

This study aimed to evaluate the performance of deep privacy-preserving federated learning (DPFL) in predicting COVID-19 outcomes using chest CT images.

METHODS

After applying inclusion and exclusion criteria, 3055 patients from 19 centers, including 1599 alive and 1456 deceased, were enrolled in this study. Data from all centers were split (randomly with stratification respective to each center and class) into a training/validation set (70%/10%) and a hold-out test set (20%). For the DL model, feature extraction was performed on 2D slices, and averaging was performed at the final layer to construct a 3D model for each scan. The DensNet model was used for feature extraction. The model was developed using centralized and FL approaches. For FL, we employed DPFL approaches. Membership inference attack was also evaluated in the FL strategy. For model evaluation, different metrics were reported in the hold-out test sets. In addition, models trained in two scenarios, centralized and FL, were compared using the DeLong test for statistical differences.

RESULTS

The centralized model achieved an accuracy of 0.76, while the DPFL model had an accuracy of 0.75. Both the centralized and DPFL models achieved a specificity of 0.77. The centralized model achieved a sensitivity of 0.74, while the DPFL model had a sensitivity of 0.73. A mean AUC of 0.82 and 0.81 with 95% confidence intervals of (95% CI: 0.79-0.85) and (95% CI: 0.77-0.84) were achieved by the centralized model and the DPFL model, respectively. The DeLong test did not prove statistically significant differences between the two models (p-value = 0.98). The AUC values for the inference attacks fluctuate between 0.49 and 0.51, with an average of 0.50 ± 0.003 and 95% CI for the mean AUC of 0.500 to 0.501.

CONCLUSION

The performance of the proposed model was comparable to centralized models while operating on large and heterogeneous multi-institutional datasets. In addition, the model was resistant to inference attacks, ensuring the privacy of shared data during the training process.

TMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images - a multi-center generalizability analysis

PURPOSE

Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach.

METHODS

Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians.

RESULTS

Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded.

CONCLUSION

TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.

Strategies for deep learning-based attenuation and scatter correction of brain 18 F-FDG PET images in the image domain

BACKGROUND

Attenuation and scatter correction is crucial for quantitative positron emission tomography (PET) imaging. Direct attenuation correction (AC) in the image domain using deep learning approaches has been recently proposed for combined PET/MR and standalone PET modalities lacking transmission scanning devices or anatomical imaging.

PURPOSE

In this study, different input settings were considered in the model training to investigate deep learning-based AC in the image space.

METHODS

Three different deep learning methods were developed for direct AC in the image space: (i) use of non-attenuation-corrected PET images as input (NonAC-PET), (ii) use of attenuation-corrected PET images with a simple two-class AC map (composed of soft-tissue and background air) obtained from NonAC-PET images (PET segmentation-based AC [SegAC-PET]), and (iii) use of both NonAC-PET and SegAC-PET images in a Double-Channel fashion to predict ground truth attenuation corrected PET images with Computed Tomography images (CTAC-PET). Since a simple two-class AC map (generated from NonAC-PET images) can easily be generated, this work assessed the added value of incorporating SegAC-PET images into direct AC in the image space. A 4-fold cross-validation scheme was adopted to train and evaluate the different models based using 80 brain 18 F-Fluorodeoxyglucose PET/CT images. The voxel-wise and region-wise accuracy of the models were examined via measuring the standardized uptake value (SUV) quantification bias in different regions of the brain.

RESULTS

The overall root mean square error (RMSE) for the Double-Channel setting was 0.157 ± 0.08 SUV in the whole brain region, while RMSEs of 0.214 ± 0.07 and 0.189 ± 0.14 SUV were observed in NonAC-PET and SegAC-PET models, respectively. A mean SUV bias of 0.01 ± 0.26% was achieved by the Double-Channel model regarding the activity concentration in cerebellum region, as opposed to 0.08 ± 0.28% and 0.05 ± 0.28% SUV biases for the network that uniquely used NonAC-PET or SegAC-PET as input, respectively. SegAC-PET images with an SUV bias of -1.15 ± 0.54%, served as a benchmark for clinically accepted errors. In general, the Double-Channel network, relying on both SegAC-PET and NonAC-PET images, outperformed the other AC models.

CONCLUSION

Since the generation of two-class AC maps from non-AC PET images is straightforward, the current study investigated the potential added value of incorporating SegAC-PET images into a deep learning-based direct AC approach. Altogether, compared with models that use only NonAC-PET and SegAC-PET images, the Double-Channel deep learning network exhibited superior attenuation correction accuracy.

The Image Biomarker Standardization Initiative: Standardized Convolutional Filters for Reproducible Radiomics and Enhanced Clinical Insights

Filters are commonly used to enhance specific structures and patterns in images, such as vessels or peritumoral regions, to enable clinical insights beyond the visible image using radiomics. However, their lack of standardization restricts reproducibility and clinical translation of radiomics decision support tools. In this special report, teams of researchers who developed radiomics software participated in a three-phase study (September 2020 to December 2022) to establish a standardized set of filters. The first two phases focused on finding reference filtered images and reference feature values for commonly used convolutional filters: mean, Laplacian of Gaussian, Laws and Gabor kernels, separable and nonseparable wavelets (including decomposed forms), and Riesz transformations. In the first phase, 15 teams used digital phantoms to establish 33 reference filtered images of 36 filter configurations. In phase 2, 11 teams used a chest CT image to derive reference values for 323 of 396 features computed from filtered images using 22 filter and image processing configurations. Reference filtered images and feature values for Riesz transformations were not established. Reproducibility of standardized convolutional filters was validated on a public data set of multimodal imaging (CT, fluorodeoxyglucose PET, and T1-weighted MRI) in 51 patients with soft-tissue sarcoma. At validation, reproducibility of 486 features computed from filtered images using nine configurations × three imaging modalities was assessed using the lower bounds of 95% CIs of intraclass correlation coefficients. Out of 486 features, 458 were found to be reproducible across nine teams with lower bounds of 95% CIs of intraclass correlation coefficients greater than 0.75. In conclusion, eight filter types were standardized with reference filtered images and reference feature values for verifying and calibrating radiomics software packages. A web-based tool is available for compliance checking.

PRIMIS: Privacy-preserving medical image sharing via deep sparsifying transform learning with obfuscation

OBJECTIVE

The primary objective of our study is to address the challenge of confidentially sharing medical images across different centers. This is often a critical necessity in both clinical and research environments, yet restrictions typically exist due to privacy concerns. Our aim is to design a privacy-preserving data-sharing mechanism that allows medical images to be stored as encoded and obfuscated representations in the public domain without revealing any useful or recoverable content from the images. In tandem, we aim to provide authorized users with compact private keys that could be used to reconstruct the corresponding images.

METHOD

Our approach involves utilizing a neural auto-encoder. The convolutional filter outputs are passed through sparsifying transformations to produce multiple compact codes. Each code is responsible for reconstructing different attributes of the image. The key privacy-preserving element in this process is obfuscation through the use of specific pseudo-random noise. When applied to the codes, it becomes computationally infeasible for an attacker to guess the correct representation for all the codes, thereby preserving the privacy of the images.

RESULTS

The proposed framework was implemented and evaluated using chest X-ray images for different medical image analysis tasks, including classification, segmentation, and texture analysis. Additionally, we thoroughly assessed the robustness of our method against various attacks using both supervised and unsupervised algorithms.

CONCLUSION

This study provides a novel, optimized, and privacy-assured data-sharing mechanism for medical images, enabling multi-party sharing in a secure manner. While we have demonstrated its effectiveness with chest X-ray images, the mechanism can be utilized in other medical images modalities as well.

Impact of simulated reduced injected dose on the assessment of amyloid PET scans

PURPOSE

To investigate the impact of reduced injected doses on the quantitative and qualitative assessment of the amyloid PET tracers [18F]flutemetamol and [18F]florbetaben.

METHODS

Cognitively impaired and unimpaired individuals (N = 250, 36% Aβ-positive) were included and injected with [18F]flutemetamol (N = 175) or [18F]florbetaben (N = 75). PET scans were acquired in list-mode (90-110 min post-injection) and reduced-dose images were simulated to generate images of 75, 50, 25, 12.5 and 5% of the original injected dose. Images were reconstructed using vendor-provided reconstruction tools and visually assessed for Aβ-pathology. SUVRs were calculated for a global cortical and three smaller regions using a cerebellar cortex reference tissue, and Centiloid was computed. Absolute and percentage differences in SUVR and CL were calculated between dose levels, and the ability to discriminate between Aβ- and Aβ + scans was evaluated using ROC analyses. Finally, intra-reader agreement between the reduced dose and 100% images was evaluated.

RESULTS

At 5% injected dose, change in SUVR was 3.72% and 3.12%, with absolute change in Centiloid 3.35CL and 4.62CL, for [18F]flutemetamol and [18F]florbetaben, respectively. At 12.5% injected dose, percentage change in SUVR and absolute change in Centiloid were < 1.5%. AUCs for discriminating Aβ- from Aβ + scans were high (AUC ≥ 0.94) across dose levels, and visual assessment showed intra-reader agreement of > 80% for both tracers.

CONCLUSION

This proof-of-concept study showed that for both [18F]flutemetamol and [18F]florbetaben, adequate quantitative and qualitative assessments can be obtained at 12.5% of the original injected dose. However, decisions to reduce the injected dose should be made considering the specific clinical or research circumstances.

Enhanced direct joint attenuation and scatter correction of whole-body PET images via context-aware deep networks

In positron emission tomography (PET), attenuation and scatter corrections are necessary steps toward accurate quantitative reconstruction of the radiopharmaceutical distribution. Inspired by recent advances in deep learning, many algorithms based on convolutional neural networks have been proposed for automatic attenuation and scatter correction, enabling applications to CT-less or MR-less PET scanners to improve performance in the presence of CT-related artifacts. A known characteristic of PET imaging is to have varying tracer uptakes for various patients and/or anatomical regions. However, existing deep learning-based algorithms utilize a fixed model across different subjects and/or anatomical regions during inference, which could result in spurious outputs. In this work, we present a novel deep learning-based framework for the direct reconstruction of attenuation and scatter-corrected PET from non-attenuation-corrected images in the absence of structural information in the inference. To deal with inter-subject and intra-subject uptake variations in PET imaging, we propose a novel model to perform subject- and region-specific filtering through modulating the convolution kernels in accordance to the contextual coherency within the neighboring slices. This way, the context-aware convolution can guide the composition of intermediate features in favor of regressing input-conditioned and/or region-specific tracer uptakes. We also utilized a large cohort of 910 whole-body studies for training and evaluation purposes, which is more than one order of magnitude larger than previous works. In our experimental studies, qualitative assessments showed that our proposed CT-free method is capable of producing corrected PET images that accurately resemble ground truth images corrected with the aid of CT scans. For quantitative assessments, we evaluated our proposed method over 112 held-out subjects and achieved an absolute relative error of 14.30±3.88% and a relative error of -2.11%±2.73% in whole-body.

Development and validation of survival prognostic models for head and neck cancer patients using machine learning and dosiomics and CT radiomics features: a multicentric study

BACKGROUND

This study aimed to investigate the value of clinical, radiomic features extracted from gross tumor volumes (GTVs) delineated on CT images, dose distributions (Dosiomics), and fusion of CT and dose distributions to predict outcomes in head and neck cancer (HNC) patients.

METHODS

A cohort of 240 HNC patients from five different centers was obtained from The Cancer Imaging Archive. Seven strategies, including four non-fusion (Clinical, CT, Dose, DualCT-Dose), and three fusion algorithms (latent low-rank representation referred (LLRR),Wavelet, weighted least square (WLS)) were applied. The fusion algorithms were used to fuse the pre-treatment CT images and 3-dimensional dose maps. Overall, 215 radiomics and Dosiomics features were extracted from the GTVs, alongside with seven clinical features incorporated. Five feature selection (FS) methods in combination with six machine learning (ML) models were implemented. The performance of the models was quantified using the concordance index (CI) in one-center-leave-out 5-fold cross-validation for overall survival (OS) prediction considering the time-to-event.

RESULTS

The mean CI and Kaplan-Meier curves were used for further comparisons. The CoxBoost ML model using the Minimal Depth (MD) FS method and the glmnet model using the Variable hunting (VH) FS method showed the best performance with CI = 0.73 ± 0.15 for features extracted from LLRR fused images. In addition, both glmnet-Cindex and Coxph-Cindex classifiers achieved a CI of 0.72 ± 0.14 by employing the dose images (+ incorporated clinical features) only.

CONCLUSION

Our results demonstrated that clinical features, Dosiomics and fusion of dose and CT images by specific ML-FS models could predict the overall survival of HNC patients with acceptable accuracy. Besides, the performance of ML methods among the three different strategies was almost comparable.

The quest for multifunctional and dedicated PET instrumentation with irregular geometries

We focus on reviewing state-of-the-art developments of dedicated PET scanners with irregular geometries and the potential of different aspects of multifunctional PET imaging. First, we discuss advances in non-conventional PET detector geometries. Then, we present innovative designs of organ-specific dedicated PET scanners for breast, brain, prostate, and cardiac imaging. We will also review challenges and possible artifacts by image reconstruction algorithms for PET scanners with irregular geometries, such as non-cylindrical and partial angular coverage geometries and how they can be addressed. Then, we attempt to address some open issues about cost/benefits analysis of dedicated PET scanners, how far are the theoretical conceptual designs from the market/clinic, and strategies to reduce fabrication cost without compromising performance.

Information fusion for fully automated segmentation of head and neck tumors from PET and CT images

BACKGROUND

PET/CT images combining anatomic and metabolic data provide complementary information that can improve clinical task performance. PET image segmentation algorithms exploiting the multi-modal information available are still lacking.

PURPOSE

Our study aimed to assess the performance of PET and CT image fusion for gross tumor volume (GTV) segmentations of head and neck cancers (HNCs) utilizing conventional, deep learning (DL), and output-level voting-based fusions.

METHODS

The current study is based on a total of 328 histologically confirmed HNCs from six different centers. The images were automatically cropped to a 200 × 200 head and neck region box, and CT and PET images were normalized for further processing. Eighteen conventional image-level fusions were implemented. In addition, a modified U2-Net architecture as DL fusion model baseline was used. Three different input, layer, and decision-level information fusions were used. Simultaneous truth and performance level estimation (STAPLE) and majority voting to merge different segmentation outputs (from PET and image-level and network-level fusions), that is, output-level information fusion (voting-based fusions) were employed. Different networks were trained in a 2D manner with a batch size of 64. Twenty percent of the dataset with stratification concerning the centers (20% in each center) were used for final result reporting. Different standard segmentation metrics and conventional PET metrics, such as SUV, were calculated.

RESULTS

In single modalities, PET had a reasonable performance with a Dice score of 0.77 ± 0.09, while CT did not perform acceptably and reached a Dice score of only 0.38 ± 0.22. Conventional fusion algorithms obtained a Dice score range of [0.76-0.81] with guided-filter-based context enhancement (GFCE) at the low-end, and anisotropic diffusion and Karhunen-Loeve transform fusion (ADF), multi-resolution singular value decomposition (MSVD), and multi-level image decomposition based on latent low-rank representation (MDLatLRR) at the high-end. All DL fusion models achieved Dice scores of 0.80. Output-level voting-based models outperformed all other models, achieving superior results with a Dice score of 0.84 for Majority_ImgFus, Majority_All, and Majority_Fast. A mean error of almost zero was achieved for all fusions using SUVpeak , SUVmean and SUVmedian .

CONCLUSION

PET/CT information fusion adds significant value to segmentation tasks, considerably outperforming PET-only and CT-only methods. In addition, both conventional image-level and DL fusions achieve competitive results. Meanwhile, output-level voting-based fusion using majority voting of several algorithms results in statistically significant improvements in the segmentation of HNC.

Differential privacy preserved federated transfer learning for multi-institutional 68Ga-PET image artefact detection and disentanglement

PURPOSE

Image artefacts continue to pose challenges in clinical molecular imaging, resulting in misdiagnoses, additional radiation doses to patients and financial costs. Mismatch and halo artefacts occur frequently in gallium-68 (68Ga)-labelled compounds whole-body PET/CT imaging. Correcting for these artefacts is not straightforward and requires algorithmic developments, given that conventional techniques have failed to address them adequately. In the current study, we employed differential privacy-preserving federated transfer learning (FTL) to manage clinical data sharing and tackle privacy issues for building centre-specific models that detect and correct artefacts present in PET images.

METHODS

Altogether, 1413 patients with 68Ga prostate-specific membrane antigen (PSMA)/DOTA-TATE (TOC) PET/CT scans from 3 countries, including 8 different centres, were enrolled in this study. CT-based attenuation and scatter correction (CT-ASC) was used in all centres for quantitative PET reconstruction. Prior to model training, an experienced nuclear medicine physician reviewed all images to ensure the use of high-quality, artefact-free PET images (421 patients' images). A deep neural network (modified U2Net) was trained on 80% of the artefact-free PET images to utilize centre-based (CeBa), centralized (CeZe) and the proposed differential privacy FTL frameworks. Quantitative analysis was performed in 20% of the clean data (with no artefacts) in each centre. A panel of two nuclear medicine physicians conducted qualitative assessment of image quality, diagnostic confidence and image artefacts in 128 patients with artefacts (256 images for CT-ASC and FTL-ASC).

RESULTS

The three approaches investigated in this study for 68Ga-PET imaging (CeBa, CeZe and FTL) resulted in a mean absolute error (MAE) of 0.42 ± 0.21 (CI 95%: 0.38 to 0.47), 0.32 ± 0.23 (CI 95%: 0.27 to 0.37) and 0.28 ± 0.15 (CI 95%: 0.25 to 0.31), respectively. Statistical analysis using the Wilcoxon test revealed significant differences between the three approaches, with FTL outperforming CeBa and CeZe (p-value < 0.05) in the clean test set. The qualitative assessment demonstrated that FTL-ASC significantly improved image quality and diagnostic confidence and decreased image artefacts, compared to CT-ASC in 68Ga-PET imaging. In addition, mismatch and halo artefacts were successfully detected and disentangled in the chest, abdomen and pelvic regions in 68Ga-PET imaging.

CONCLUSION

The proposed approach benefits from using large datasets from multiple centres while preserving patient privacy. Qualitative assessment by nuclear medicine physicians showed that the proposed model correctly addressed two main challenging artefacts in 68Ga-PET imaging. This technique could be integrated in the clinic for 68Ga-PET imaging artefact detection and disentanglement using multicentric heterogeneous datasets.

Tensor radiomics: paradigm for systematic incorporation of multi-flavoured radiomics features

Background

Radiomics features hold significant value as quantitative imaging biomarkers for diagnosis, prognosis, and treatment response assessment. To generate radiomics features and ultimately develop signatures, various factors can be manipulated, including image discretization parameters (e.g., bin number or size), convolutional filters, segmentation perturbation, or multi-modality fusion levels. Typically, only one set of parameters is employed, resulting in a single value or "flavour" for each radiomics feature. In contrast, we propose "tensor radiomics" (TR) where tensors of features calculated using multiple parameter combinations (i.e., flavours) are utilized to optimize the creation of radiomics signatures.

Methods

We provide illustrative instances of TR implementation in positron emission tomography-computed tomography (PET-CT), magnetic resonance imaging (MRI), and CT by leveraging machine learning (ML) and deep learning (DL) methodologies, as well as reproducibility analyses: (I) to predict overall survival (OS) in lung cancer (CT) and head and neck cancer (PET-CT), TR was employed by varying bin sizes. This approach involved use of a hybrid deep neural network called 'TR-Net' and two ML-based techniques for combining different flavours. (II) TR was constructed by incorporating different segmentation perturbations and various bin sizes to classify the response of late-stage lung cancer to first-line immunotherapy using CT images. (III) In MRI of glioblastoma (GBM), TR was implemented to generate multi-flavour radiomics features, enabling enhanced analysis and interpretation. (IV) TR was employed via multiple PET-CT fusions in head and neck cancer. Flavours based on different fusions were created using Laplacian pyramids and wavelet transforms.

Results

Our findings demonstrated that TR outperformed conventional radiomics features in lung cancer CT and head and neck cancer PET-CT images, significantly enhancing OS prediction accuracy. TR also improved classification of lung cancer response to therapy and exhibited notable advantages in reproducibility compared to single-flavour features in MR imaging of GBM. Moreover, in head and neck cancer, TR through multiple PET-CT fusions exhibited improved performance in predicting OS.

Conclusions

We conclude that the proposed TR paradigm has significant potential to improve performance in different medical imaging tasks. By incorporating multiple flavours of radiomics features, TR overcomes limitations associated with individual features and shows promise in enhancing prognostic capabilities in clinical settings.

Left Ventricular Myocardial Dysfunction Evaluation in Thalassemia Patients Using Echocardiographic Radiomic Features and Machine Learning Algorithms

Heart failure caused by iron deposits in the myocardium is the primary cause of mortality in beta-thalassemia major patients. Cardiac magnetic resonance imaging (CMRI) T2* is the primary screening technique used to detect myocardial iron overload, but inherently bears some limitations. In this study, we aimed to differentiate beta-thalassemia major patients with myocardial iron overload from those without myocardial iron overload (detected by T2*CMRI) based on radiomic features extracted from echocardiography images and machine learning (ML) in patients with normal left ventricular ejection fraction (LVEF > 55%) in echocardiography. Out of 91 cases, 44 patients with thalassemia major with normal LVEF (> 55%) and T2* ≤ 20 ms and 47 people with LVEF > 55% and T2* > 20 ms as the control group were included in the study. Radiomic features were extracted for each end-systolic (ES) and end-diastolic (ED) image. Then, three feature selection (FS) methods and six different classifiers were used. The models were evaluated using various metrics, including the area under the ROC curve (AUC), accuracy (ACC), sensitivity (SEN), and specificity (SPE). Maximum relevance-minimum redundancy-eXtreme gradient boosting (MRMR-XGB) (AUC = 0.73, ACC = 0.73, SPE = 0.73, SEN = 0.73), ANOVA-MLP (AUC = 0.69, ACC = 0.69, SPE = 0.56, SEN = 0.83), and recursive feature elimination-K-nearest neighbors (RFE-KNN) (AUC = 0.65, ACC = 0.65, SPE = 0.64, SEN = 0.65) were the best models in ED, ES, and ED&ES datasets. Using radiomic features extracted from echocardiographic images and ML, it is feasible to predict cardiac problems caused by iron overload.

Real-time, acquisition parameter-free voxel-wise patient-specific Monte Carlo dose reconstruction in whole-body CT scanning using deep neural networks

OBJECTIVE

We propose a deep learning-guided approach to generate voxel-based absorbed dose maps from whole-body CT acquisitions.

METHODS

The voxel-wise dose maps corresponding to each source position/angle were calculated using Monte Carlo (MC) simulations considering patient- and scanner-specific characteristics (SP_MC). The dose distribution in a uniform cylinder was computed through MC calculations (SP_uniform). The density map and SP_uniform dose maps were fed into a residual deep neural network (DNN) to predict SP_MC through an image regression task. The whole-body dose maps reconstructed by the DNN and MC were compared in the 11 test cases scanned with two tube voltages through transfer learning with/without tube current modulation (TCM). The voxel-wise and organ-wise dose evaluations, such as mean error (ME, mGy), mean absolute error (MAE, mGy), relative error (RE, %), and relative absolute error (RAE, %), were performed.

RESULTS

The model performance for the 120 kVp and TCM test set in terms of ME, MAE, RE, and RAE voxel-wise parameters was  - 0.0302 ± 0.0244 mGy, 0.0854 ± 0.0279 mGy,  - 1.13 ± 1.41%, and 7.17 ± 0.44%, respectively. The organ-wise errors for 120 kVp and TCM scenario averaged over all segmented organs in terms of ME, MAE, RE, and RAE were  - 0.144 ± 0.342 mGy, and 0.23 ± 0.28 mGy,  - 1.11 ± 2.90%, 2.34 ± 2.03%, respectively.

CONCLUSION

Our proposed deep learning model is able to generate voxel-level dose maps from a whole-body CT scan with reasonable accuracy suitable for organ-level absorbed dose estimation.

CLINICAL RELEVANCE STATEMENT

We proposed a novel method for voxel dose map calculation using deep neural networks. This work is clinically relevant since accurate dose calculation for patients can be carried out within acceptable computational time compared to lengthy Monte Carlo calculations.

KEY POINTS

• We proposed a deep neural network approach as an alternative to Monte Carlo dose calculation. • Our proposed deep learning model is able to generate voxel-level dose maps from a whole-body CT scan with reasonable accuracy, suitable for organ-level dose estimation. • By generating a dose distribution from a single source position, our model can generate accurate and personalized dose maps for a wide range of acquisition parameters.

Time-to-event overall survival prediction in glioblastoma multiforme patients using magnetic resonance imaging radiomics

PURPOSE

Glioblastoma Multiforme (GBM) represents the predominant aggressive primary tumor of the brain with short overall survival (OS) time. We aim to assess the potential of radiomic features in predicting the time-to-event OS of patients with GBM using machine learning (ML) algorithms.

MATERIALS AND METHODS

One hundred nineteen patients with GBM, who had T1-weighted contrast-enhanced and T2-FLAIR MRI sequences, along with clinical data and survival time, were enrolled. Image preprocessing methods included 64 bin discretization, Laplacian of Gaussian (LOG) filters with three Sigma values and eight variations of Wavelet Transform. Images were then segmented, followed by the extraction of 1212 radiomic features. Seven feature selection (FS) methods and six time-to-event ML algorithms were utilized. The combination of preprocessing, FS, and ML algorithms (12 × 7 × 6 = 504 models) was evaluated by multivariate analysis.

RESULTS

Our multivariate analysis showed that the best prognostic FS/ML combinations are the Mutual Information (MI)/Cox Boost, MI/Generalized Linear Model Boosting (GLMB) and MI/Generalized Linear Model Network (GLMN), all of which were done via the LOG (Sigma = 1 mm) preprocessing method (C-index = 0.77). The LOG filter with Sigma = 1 mm preprocessing method, MI, GLMB and GLMN achieved significantly higher C-indices than other preprocessing, FS, and ML methods (all p values < 0.05, mean C-indices of 0.65, 0.70, and 0.64, respectively).

CONCLUSION

ML algorithms are capable of predicting the time-to-event OS of patients using MRI-based radiomic and clinical features. MRI-based radiomics analysis in combination with clinical variables might appear promising in assisting clinicians in the survival prediction of patients with GBM. Further research is needed to establish the applicability of radiomics in the management of GBM in the clinic.

Artificial Intelligence-Driven Single-Shot PET Image Artifact Detection and Disentanglement: Toward Routine Clinical Image Quality Assurance

PURPOSE

Medical imaging artifacts compromise image quality and quantitative analysis and might confound interpretation and misguide clinical decision-making. The present work envisions and demonstrates a new paradigm PET image Quality Assurance NETwork (PET-QA-NET) in which various image artifacts are detected and disentangled from images without prior knowledge of a standard of reference or ground truth for routine PET image quality assurance.

METHODS

The network was trained and evaluated using training/validation/testing data sets consisting of 669/100/100 artifact-free oncological 18 F-FDG PET/CT images and subsequently fine-tuned and evaluated on 384 (20% for fine-tuning) scans from 8 different PET centers. The developed DL model was quantitatively assessed using various image quality metrics calculated for 22 volumes of interest defined on each scan. In addition, 200 additional 18 F-FDG PET/CT scans (this time with artifacts), generated using both CT-based attenuation and scatter correction (routine PET) and PET-QA-NET, were blindly evaluated by 2 nuclear medicine physicians for the presence of artifacts, diagnostic confidence, image quality, and the number of lesions detected in different body regions.

RESULTS

Across the volumes of interest of 100 patients, SUV MAE values of 0.13 ± 0.04, 0.24 ± 0.1, and 0.21 ± 0.06 were reached for SUV mean , SUV max , and SUV peak , respectively (no statistically significant difference). Qualitative assessment showed a general trend of improved image quality and diagnostic confidence and reduced image artifacts for PET-QA-NET compared with routine CT-based attenuation and scatter correction.

CONCLUSION

We developed a highly effective and reliable quality assurance tool that can be embedded routinely to detect and correct for 18 F-FDG PET image artifacts in clinical setting with notably improved PET image quality and quantitative capabilities.

Accurate modeling and performance evaluation of a total-body pet scanner using Monte Carlo simulations

BACKGROUND

The limited axial field-of-view (FOV) of conventional clinical positron emission tomography (PET) scanners (∼15 to 26 cm) allows detecting only 1% of all coincidence photons, hence limiting significantly their sensitivity. To overcome this limitation, the EXPLORER consortium developed the world's first total-body PET/CT scanner that significantly increased the sensitivity, thus enabling to decrease the scan duration or injected dose.

PURPOSE

The purpose of this study is to perform and validate Monte Carlo simulations of the uEXPLORER PET scanner, which can be used to devise novel conceptual designs and geometrical configurations through obtaining features that are difficult to obtain experimentally.

METHODS

The total-body uEXPLORER PET scanner was modeled using GATE Monte Carlo (MC) platform. The model was validated through comparison with experimental measurements of various performance parameters, including spatial resolution, sensitivity, count rate performance, and image quality, according to NEMA-NU2 2018 standards. Furthermore, the effects of the time coincidence window and maximum ring difference on the count rate and noise equivalent count rate (NECR) were evaluated.

RESULTS

Overall, the validation study showed that there was a good agreement between the simulation and experimental results. The differences between the simulated and experimental total sensitivity for the NEMA and extended phantoms at the center of the FOV were 2.3% and 0.0%, respectively. The difference in peak NECR was 9.9% for the NEMA phantom and 1.0% for the extended phantom. The average bias between the simulated and experimental results of the full-width-at-half maximum (FWHM) for six different positions and three directions was 0.12 mm. The simulations showed that using a variable coincidence time window based on the maximum ring difference can reduce the effect of random coincidences and improve the NECR compared to a constant time coincidence window. The NECR corresponding to 252-ring difference was 2.11 Mcps, which is larger than the NECR corresponding to 336-ring difference (2.04 Mcps).

CONCLUSION

The developed MC model of the uEXPLORER PET scanner was validated against experimental measurements and can be used for further assessment and design optimization of the scanner.

Development and validation of an optimal GATE model for proton pencil-beam scanning delivery

OBJECTIVE

To develop and validate a versatile Monte Carlo (MC)-based dose calculation engine to support MC-based dose verification of treatment planning systems (TPSs) and quality assurance (QA) workflows in proton therapy.

METHODS

The GATE MC toolkit was used to simulate a fixed horizontal active scan-based proton beam delivery (SIEMENS IONTRIS). Within the nozzle, two primary and secondary dose monitors have been designed to enable the comparison of the accuracy of dose estimation from MC simulations with respect to physical QA measurements. The developed beam model was validated against a series of commissioning measurements using pinpoint chambers and 2D array ionization chambers (IC) in terms of lateral profiles and depth dose distributions. Furthermore, beam delivery module and treatment planning has been validated against the literature deploying various clinical test cases of the AAPM TG-119 (c-shape phantom) and a prostate patient.

RESULTS

MC simulations showed excellent agreement with measurements in the lateral depth-dose parameters and spread-out Bragg peak (SOBP) characteristics within a maximum relative error of 0.95 mm in range, 1.83% in entrance to peak ratio, 0.27% in mean point-to-point dose difference, and 0.32% in peak location. The mean relative absolute difference between MC simulations and measurements in terms of absorbed dose in the SOBP region was 0.93% ± 0.88%. Clinical phantom studies showed a good agreement compared to research TPS (relative error for TG-119 planning target volume PTV-D95 ∼ 1.8%; and for prostate PTV-D95 ∼ -0.6%).

CONCLUSION

We successfully developed a MC model for the pencil beam scanning system, which appears reliable for dose verification of the TPS in combination with QA information, prior to patient treatment.

Dual-Centre Harmonised Multimodal Positron Emission Tomography/Computed Tomography Image Radiomic Features and Machine Learning Algorithms for Non-small Cell Lung Cancer Histopathological Subtype Phenotype Decoding

AIMS

We aimed to build radiomic models for classifying non-small cell lung cancer (NSCLC) histopathological subtypes through a dual-centre dataset and comprehensively evaluate the effect of ComBat harmonisation on the performance of single- and multimodality radiomic models.

MATERIALS AND METHODS

A public dataset of NSCLC patients from two independent centres was used. Two image fusion methods, namely guided filtering-based fusion and image fusion based on visual saliency map and weighted least square optimisation, were used. Radiomic features were extracted from each scan, including first-order, texture and moment-invariant features. Subsequently, ComBat harmonisation was applied to the extracted features from computed tomography (CT), positron emission tomography (PET) and fused images to correct the centre effect. For feature selection, least absolute shrinkage and selection operator (Lasso) and recursive feature elimination (RFE) were investigated. For machine learning, logistic regression (LR), support vector machine (SVM) and AdaBoost were evaluated for classifying NSCLC subtypes. Training and evaluation of the models were carried out in a robust framework to offset plausible errors and performance was reported using area under the curve, balanced accuracy, sensitivity and specificity before and after harmonisation. N-way ANOVA was used to assess the effect of different factors on the performance of the models.

RESULTS

Support vector machine fed with selected features by recursive feature elimination from a harmonised PET feature set achieved the highest performance (area under the curve = 0.82) in classifying NSCLC histopathological subtypes. Although the performance of the models did not significantly improve for CT images after harmonisation, the performance of PET and guided filtering-based fusion feature signatures significantly improved for almost all models. Although the selection of the image modality and feature selection methods was effective on the performance of the model (ANOVA P-values <0.001), machine learning and harmonisation did not change the performance significantly (ANOVA P-values = 0.839 and 0.292, respectively).

CONCLUSION

This study confirmed the potential of radiomic analysis on PET, CT and hybrid images for histopathological classification of NSCLC subtypes.

Recent Advances in Positron Emission Tomography/Magnetic Resonance Imaging Technology

More than a decade has passed since the clinical deployment of the first commercial whole-body hybrid PET/MR scanner in the clinic. The major advantages and limitations of this technology have been investigated from technical and medical perspectives. Despite the remarkable advantages associated with hybrid PET/MR imaging, such as reduced radiation dose and fully simultaneous functional and structural imaging, this technology faced major challenges in terms of mutual interference between MRI and PET components, in addition to the complexity of achieving quantitative imaging owing to the intricate MRI-guided attenuation correction in PET/MRI. In this review, the latest technical developments in PET/MRI technology as well as the state-of-the-art solutions to the major challenges of quantitative PET/MR imaging are discussed.

PET/MR Imaging in Head and Neck Cancer

Head and neck squamous cell carcinoma (HNSCC) can either be examined with hybrid PET/MR imaging systems or sequentially, using PET/CT and MR imaging. Regardless of the acquisition technique, the superiority of MR imaging compared to CT lies in its potential to interrogate tumor and surrounding tissues with different sequences, including perfusion and diffusion. For this reason, PET/MR imaging is preferable for the detection and assessment of locoregional residual/recurrent HNSCC after therapy. In addition, MR imaging interpretation is facilitated when combined with PET. Nevertheless, distant metastases and distant second primary tumors are detected equally well with PET/MR imaging and PET/CT.

Multi-institutional PET/CT image segmentation using federated deep transformer learning

BACKGROUND AND OBJECTIVE

Generalizable and trustworthy deep learning models for PET/CT image segmentation necessitates large diverse multi-institutional datasets. However, legal, ethical, and patient privacy issues challenge sharing of datasets between different centers. To overcome these challenges, we developed a federated learning (FL) framework for multi-institutional PET/CT image segmentation.

METHODS

A dataset consisting of 328 FL (HN) cancer patients who underwent clinical PET/CT examinations gathered from six different centers was enrolled. A pure transformer network was implemented as fully core segmentation algorithms using dual channel PET/CT images. We evaluated different frameworks (single center-based, centralized baseline, as well as seven different FL algorithms) using 68 PET/CT images (20% of each center data). In particular, the implemented FL algorithms include clipping with the quantile estimator (ClQu), zeroing with the quantile estimator (ZeQu), federated averaging (FedAvg), lossy compression (LoCo), robust aggregation (RoAg), secure aggregation (SeAg), and Gaussian differentially private FedAvg with adaptive quantile clipping (GDP-AQuCl).

RESULTS

The Dice coefficient was 0.80±0.11 for both centralized and SeAg FL algorithms. All FL approaches achieved centralized learning model performance with no statistically significant differences. Among the FL algorithms, SeAg and GDP-AQuCl performed better than the other techniques. However, there was no statistically significant difference. All algorithms, except the center-based approach, resulted in relative errors less than 5% for SUVmax and SUVmean for all FL and centralized methods. Centralized and FL algorithms significantly outperformed the single center-based baseline.

CONCLUSIONS

The developed FL-based (with centralized method performance) algorithms exhibited promising performance for HN tumor segmentation from PET/CT images.

Hydrogen nanobubbles: A novel approach toward radio-sensitization agents

BACKGROUND

Ocular melanoma is a rare kind of eye malignancy that threatens the patient's eyesight. Radiotherapy and surgical removal are the most commonly used therapeutic modalities, and nanomedicine has lately entered this field. Brachytherapy using Ruthenium-106 (106 Ru) ophthalmic plaques has been used for decades to treat ocular melanoma, with the applicator placed on the patient's eyes until the prescribed dose reaches the tumor apex.

PURPOSE

To investigate the efficiency of hydrogen nanobubbles (H2 -NBs) employment during intraocular melanoma brachytherapy using a 106 Ru electron emitter plaque.

METHODS

The Monte Carlo (MC) simulation and experimental investigation using a 3D-designed phantom and thermoluminescence dosimetry (TLD) were employed. Various concentrations of H2 -NBs with a diameter of 100 nm were simulated inside tumor tissue. The results were presented as deposited energy and dose enhancement factor (DEF). An equivalent Resin phantom of the human eyeball was made using AutoCAD and 3D-Printer technologies. The glass-bead TLDs dosimeter were employed and placed inside the phantom.

RESULTS

Using a 1% concentration of H2 -NBs, a DEF of 93% and 98% were achieved at the tumor apex of 10 mm from the experimental setup and MC simulation, respectively. For simulated concentrations of 0.1%, 0.3%, 0.5%, 1%, and 4% H2 -NBs, a maximum dose enhancement of 154%, 174%, 188%, 200%, and 300% were achieved, respectively, and a dose reduction was seen at about 3 mm from the plaque surface.

CONCLUSION

H2 -NBs can be used as an absorbed dose enhancer in 106 Ru eye brachytherapy because of their unique physical characteristics. Reducing plaque implantation time on the patient's eye, reducing sclera absorbed dose, and decreasing the risk of patients' healthy organs irradiation are reported as some of the potential benefits of using H2-NBs.

Utility of a simplified [18F] sodium fluoride PET imaging method to quantify bone metabolic flux for a wide range of clinical applications

We review the rationale, methodology, and clinical utility of quantitative [18F] sodium fluoride ([18F]NaF) positron emission tomography-computed tomography (PET-CT) imaging to measure bone metabolic flux (Ki, also known as bone plasma clearance), a measurement indicative of the local rate of bone formation at the chosen region of interest. We review the bone remodelling cycle and explain what aspects of bone remodelling are addressed by [18F]NaF PET-CT. We explain how the technique works, what measurements are involved, and what makes [18F]NaF PET-CT a useful tool for the study of bone remodelling. We discuss how these measurements can be simplified without loss of accuracy to make the technique more accessible. Finally, we briefly review some key clinical applications and discuss the potential for future developments. We hope that the simplified method described here will assist in promoting the wider use of the technique.

Machine learning-based diagnosis and risk classification of coronary artery disease using myocardial perfusion imaging SPECT: A radiomics study

This study aimed to investigate the diagnostic performance of machine learning-based radiomics analysis to diagnose coronary artery disease status and risk from rest/stress Myocardial Perfusion Imaging (MPI) single-photon emission computed tomography (SPECT). A total of 395 patients suspicious of coronary artery disease who underwent 2-day stress-rest protocol MPI SPECT were enrolled in this study. The left ventricle myocardium, excluding the cardiac cavity, was manually delineated on rest and stress images to define a volume of interest. Added to clinical features (age, sex, family history, diabetes status, smoking, and ejection fraction), a total of 118 radiomics features, were extracted from rest and stress MPI SPECT images to establish different feature sets, including Rest-, Stress-, Delta-, and Combined-radiomics (all together) feature sets. The data were randomly divided into 80% and 20% subsets for training and testing, respectively. The performance of classifiers built from combinations of three feature selections, and nine machine learning algorithms was evaluated for two different diagnostic tasks, including 1) normal/abnormal (no CAD vs. CAD) classification, and 2) low-risk/high-risk CAD classification. Different metrics, including the area under the ROC curve (AUC), accuracy (ACC), sensitivity (SEN), and specificity (SPE), were reported for models' evaluation. Overall, models built on the Stress feature set (compared to other feature sets), and models to diagnose the second task (compared to task 1 models) revealed better performance. The Stress-mRMR-KNN (feature set-feature selection-classifier) reached the highest performance for task 1 with AUC, ACC, SEN, and SPE equal to 0.61, 0.63, 0.64, and 0.6, respectively. The Stress-Boruta-GB model achieved the highest performance for task 2 with AUC, ACC, SEN, and SPE of 0.79, 0.76, 0.75, and 0.76, respectively. Diabetes status from the clinical feature family, and dependence count non-uniformity normalized, from the NGLDM family, which is representative of non-uniformity in the region of interest were the most frequently selected features from stress feature set for CAD risk classification. This study revealed promising results for CAD risk classification using machine learning models built on MPI SPECT radiomics. The proposed models are helpful to alleviate the labor-intensive MPI SPECT interpretation process regarding CAD status and can potentially expedite the diagnostic process.

An active learning approach to train a deep learning algorithm for tumor segmentation from brain MR images

PURPOSE

This study focuses on assessing the performance of active learning techniques to train a brain MRI glioma segmentation model.

METHODS

The publicly available training dataset provided for the 2021 RSNA-ASNR-MICCAI Brain Tumor Segmentation (BraTS) Challenge was used in this study, consisting of 1251 multi-institutional, multi-parametric MR images. Post-contrast T1, T2, and T2 FLAIR images as well as ground truth manual segmentation were used as input for the model. The data were split into a training set of 1151 cases and testing set of 100 cases, with the testing set remaining constant throughout. Deep convolutional neural network segmentation models were trained using the NiftyNet platform. To test the viability of active learning in training a segmentation model, an initial reference model was trained using all 1151 training cases followed by two additional models using only 575 cases and 100 cases. The resulting predicted segmentations of these two additional models on the remaining training cases were then addended to the training dataset for additional training.

RESULTS

It was demonstrated that an active learning approach for manual segmentation can lead to comparable model performance for segmentation of brain gliomas (0.906 reference Dice score vs 0.868 active learning Dice score) while only requiring manual annotation for 28.6% of the data.

CONCLUSION

The active learning approach when applied to model training can drastically reduce the time and labor spent on preparation of ground truth training data.

CRITICAL RELEVANCE STATEMENT

Active learning concepts were applied to a deep learning-assisted segmentation of brain gliomas from MR images to assess their viability in reducing the required amount of manually annotated ground truth data in model training.

KEY POINTS

• This study focuses on assessing the performance of active learning techniques to train a brain MRI glioma segmentation model. • The active learning approach for manual segmentation can lead to comparable model performance for segmentation of brain gliomas. • Active learning when applied to model training can drastically reduce the time and labor spent on preparation of ground truth training data.

Post-revascularization Ejection Fraction Prediction for Patients Undergoing Percutaneous Coronary Intervention Based on Myocardial Perfusion SPECT Imaging Radiomics: a Preliminary Machine Learning Study

In this study, the ability of radiomics features extracted from myocardial perfusion imaging with SPECT (MPI-SPECT) was investigated for the prediction of ejection fraction (EF) post-percutaneous coronary intervention (PCI) treatment. A total of 52 patients who had undergone pre-PCI MPI-SPECT were enrolled in this study. After normalization of the images, features were extracted from the left ventricle, initially automatically segmented by k-means and active contour methods, and finally edited and approved by an expert radiologist. More than 1700 2D and 3D radiomics features were extracted from each patient's scan. A cross-combination of three feature selections and seven classifier methods was implemented. Three classes of no or dis-improvement (class 1), improved EF from 0 to 5% (class 2), and improved EF over 5% (class 3) were predicted by using tenfold cross-validation. Lastly, the models were evaluated based on accuracy, AUC, sensitivity, specificity, precision, and F-score. Neighborhood component analysis (NCA) selected the most predictive feature signatures, including Gabor, first-order, and NGTDM features. Among the classifiers, the best performance was achieved by the fine KNN classifier, which yielded mean accuracy, AUC, sensitivity, specificity, precision, and F-score of 0.84, 0.83, 0.75, 0.87, 0.78, and 0.76, respectively, in 100 iterations of classification, within the 52 patients with 10-fold cross-validation. The MPI-SPECT-based radiomic features are well suited for predicting post-revascularization EF and therefore provide a helpful approach for deciding on the most appropriate treatment.

The predictive value of pretherapy [68Ga]Ga-DOTA-TATE PET and biomarkers in [177Lu]Lu-PRRT tumor dosimetry

PURPOSE

Metastatic neuroendocrine tumors (NETs) overexpressing type 2 somatostatin receptors are the target for peptide receptor radionuclide therapy (PRRT) through the theragnostic pair of 68Ga/177Lu-DOTATATE. The main purpose of this study was to develop machine learning models to predict therapeutic tumor dose using pre therapy 68Ga -PET and clinicopathological biomarkers.

METHODS

We retrospectively analyzed 90 segmented metastatic NETs from 25 patients (M14/F11, age 63.7 ± 9.5, range 38-76) treated by 177Lu-DOTATATE at our institute. Patients underwent both pretherapy [68Ga]Ga-DOTA-TATE PET/CT and four timepoints SPECT/CT at  ~ 4, 24, 96, and 168 h post-177Lu-DOTATATE infusion. Tumors were segmented by a radiologist on baseline CT or MRI and transferred to co-registered PET/CT and SPECT/CT, and normal organs were segmented by deep learning-based method on CT of the PET and SPECT. The SUV metrics and tumor-to-normal tissue SUV ratios (SUV_TNRs) were calculated from 68Ga -PET at the contour-level. Posttherapy dosimetry was performed based on the co-registration of SPECT/CTs to generate time-integrated-activity, followed by an in-house Monte Carlo-based absorbed dose estimation. The correlation between delivered 177Lu Tumor absorbed dose and PET-derived metrics along with baseline clinicopathological biomarkers (such as Creatinine, Chromogranin A and prior therapies) were evaluated. Multiple interpretable machine-learning algorithms were developed to predict tumor dose using these pretherapy information. Model performance on a nested tenfold cross-validation was evaluated in terms of coefficient of determination (R2), mean-absolute-error (MAE), and mean-relative-absolute-error (MRAE).

RESULTS

SUVmean showed a significant correlation (q-value < 0.05) with absorbed dose (Spearman ρ = 0.64), followed by TLSUVmean (SUVmean of total-lesion-burden) and SUVpeak (ρ = 0.45 and 0.41, respectively). The predictive value of PET-SUVmean in estimation of posttherapy absorbed dose was stronger compared to PET-SUVpeak, and SUV_TNRs in terms of univariate analysis (R2 = 0.28 vs. R2 ≤ 0.12). An optimal trivariate random forest model composed of SUVmean, TLSUVmean, and total liver SUVmean (normal and tumoral liver) provided the best performance in tumor dose prediction with R2 = 0.64, MAE = 0.73 Gy/GBq, and MRAE = 0.2.

CONCLUSION

Our preliminary results demonstrate the feasibility of using baseline PET images for prediction of absorbed dose prior to 177Lu-PRRT. Machine learning models combining multiple PET-based metrics performed better than using a single SUV value and using other investigated clinicopathological biomarkers. Developing such quantitative models forms the groundwork for the role of 68Ga -PET not only for the implementation of personalized treatment planning but also for patient stratification in the era of precision medicine.

Reduction of scan duration and radiation dose in cerebral CT perfusion imaging of acute stroke using a recurrent neural network

Objective. Cerebral CT perfusion (CTP) imaging is most commonly used to diagnose acute ischaemic stroke and support treatment decisions. Shortening CTP scan duration is desirable to reduce the accumulated radiation dose and the risk of patient head movement. In this study, we present a novel application of a stochastic adversarial video prediction approach to reduce CTP imaging acquisition time.Approach. A variational autoencoder and generative adversarial network (VAE-GAN) were implemented in a recurrent framework in three scenarios: to predict the last 8 (24 s), 13 (31.5 s) and 18 (39 s) image frames of the CTP acquisition from the first 25 (36 s), 20 (28.5 s) and 15 (21 s) acquired frames, respectively. The model was trained using 65 stroke cases and tested on 10 unseen cases. Predicted frames were assessed against ground-truth in terms of image quality and haemodynamic maps, bolus shape characteristics and volumetric analysis of lesions.Main results. In all three prediction scenarios, the mean percentage error between the area, full-width-at-half-maximum and maximum enhancement of the predicted and ground-truth bolus curve was less than 4 ± 4%. The best peak signal-to-noise ratio and structural similarity of predicted haemodynamic maps was obtained for cerebral blood volume followed (in order) by cerebral blood flow, mean transit time and time to peak. For the 3 prediction scenarios, average volumetric error of the lesion was overestimated by 7%-15%, 11%-28% and 7%-22% for the infarct, penumbra and hypo-perfused regions, respectively, and the corresponding spatial agreement for these regions was 67%-76%, 76%-86% and 83%-92%.Significance. This study suggests that a recurrent VAE-GAN could potentially be used to predict a portion of CTP frames from truncated acquisitions, preserving the majority of clinical content in the images, and potentially reducing the scan duration and radiation dose simultaneously by 65% and 54.5%, respectively.

Predictive value of 99mTc-MAA-based dosimetry in personalized 90Y-SIRT planning for liver malignancies

BACKGROUND

Selective internal radiation therapy with ⁹⁰Y radioembolization aims to selectively irradiate liver tumours by administering radioactive microspheres under the theragnostic assumption that the pre-therapy injection of ^99mTc labelled macroaggregated albumin (^99mTc-MAA) provides an estimation of the ⁹⁰Y microspheres biodistribution, which is not always the case. Due to the growing interest in theragnostic dosimetry for personalized radionuclide therapy, a robust relationship between the delivered and pre-treatment radiation absorbed doses is required. In this work, we aim to investigate the predictive value of absorbed dose metrics calculated from ^99mTc-MAA (simulation) compared to those obtained from ⁹⁰Y post-therapy SPECT/CT.

RESULTS

A total of 79 patients were analysed. Pre- and post-therapy 3D-voxel dosimetry was calculated on ^99mTc-MAA and ⁹⁰Y SPECT/CT, respectively, based on Local Deposition Method. Mean absorbed dose, tumour-to-normal ratio, and absorbed dose distribution in terms of dose-volume histogram (DVH) metrics were obtained and compared for each volume of interest (VOI). Mann-Whitney U-test and Pearson's correlation coefficient were used to assess the correlation between both methods. The effect of the tumoral liver volume on the absorbed dose metrics was also investigated. Strong correlation was found between simulation and therapy mean absorbed doses for all VOIs, although simulation tended to overestimate tumour absorbed doses by 26%. DVH metrics showed good correlation too, but significant differences were found for several metrics, mostly on non-tumoral liver. It was observed that the tumoral liver volume does not significantly affect the differences between simulation and therapy absorbed dose metrics.

CONCLUSION

This study supports the strong correlation between absorbed dose metrics from simulation and therapy dosimetry based on ⁹⁰Y SPECT/CT, highlighting the predictive ability of ^99mTc-MAA, not only in terms of mean absorbed dose but also of the dose distribution.

A cycle-consistent adversarial network for brain PET partial volume correction without prior anatomical information

PURPOSE

Partial volume effect (PVE) is a consequence of the limited spatial resolution of PET scanners. PVE can cause the intensity values of a particular voxel to be underestimated or overestimated due to the effect of surrounding tracer uptake. We propose a novel partial volume correction (PVC) technique to overcome the adverse effects of PVE on PET images.

METHODS

Two hundred and twelve clinical brain PET scans, including 50 18F-Fluorodeoxyglucose (18F-FDG), 50 18F-Flortaucipir, 36 18F-Flutemetamol, and 76 18F-FluoroDOPA, and their corresponding T1-weighted MR images were enrolled in this study. The Iterative Yang technique was used for PVC as a reference or surrogate of the ground truth for evaluation. A cycle-consistent adversarial network (CycleGAN) was trained to directly map non-PVC PET images to PVC PET images. Quantitative analysis using various metrics, including structural similarity index (SSIM), root mean squared error (RMSE), and peak signal-to-noise ratio (PSNR), was performed. Furthermore, voxel-wise and region-wise-based correlations of activity concentration between the predicted and reference images were evaluated through joint histogram and Bland and Altman analysis. In addition, radiomic analysis was performed by calculating 20 radiomic features within 83 brain regions. Finally, a voxel-wise two-sample t-test was used to compare the predicted PVC PET images with reference PVC images for each radiotracer.

RESULTS

The Bland and Altman analysis showed the largest and smallest variance for 18F-FDG (95% CI: - 0.29, + 0.33 SUV, mean = 0.02 SUV) and 18F-Flutemetamol (95% CI: - 0.26, + 0.24 SUV, mean =  - 0.01 SUV), respectively. The PSNR was lowest (29.64 ± 1.13 dB) for 18F-FDG and highest (36.01 ± 3.26 dB) for 18F-Flutemetamol. The smallest and largest SSIM were achieved for 18F-FDG (0.93 ± 0.01) and 18F-Flutemetamol (0.97 ± 0.01), respectively. The average relative error for the kurtosis radiomic feature was 3.32%, 9.39%, 4.17%, and 4.55%, while it was 4.74%, 8.80%, 7.27%, and 6.81% for NGLDM_contrast feature for 18F-Flutemetamol, 18F-FluoroDOPA, 18F-FDG, and 18F-Flortaucipir, respectively.

CONCLUSION

An end-to-end CycleGAN PVC method was developed and evaluated. Our model generates PVC images from the original non-PVC PET images without requiring additional anatomical information, such as MRI or CT. Our model eliminates the need for accurate registration or segmentation or PET scanner system response characterization. In addition, no assumptions regarding anatomical structure size, homogeneity, boundary, or background level are required.

Myocardial Perfusion SPECT Imaging Radiomic Features and Machine Learning Algorithms for Cardiac Contractile Pattern Recognition

A U-shaped contraction pattern was shown to be associated with a better Cardiac resynchronization therapy (CRT) response. The main goal of this study is to automatically recognize left ventricular contractile patterns using machine learning algorithms trained on conventional quantitative features (ConQuaFea) and radiomic features extracted from Gated single-photon emission computed tomography myocardial perfusion imaging (GSPECT MPI). Among 98 patients with standard resting GSPECT MPI included in this study, 29 received CRT therapy and 69 did not (also had CRT inclusion criteria but did not receive treatment yet at the time of data collection, or refused treatment). A total of 69 non-CRT patients were employed for training, and the 29 were employed for testing. The models were built utilizing features from three distinct feature sets (ConQuaFea, radiomics, and ConQuaFea + radiomics (combined)), which were chosen using Recursive feature elimination (RFE) feature selection (FS), and then trained using seven different machine learning (ML) classifiers. In addition, CRT outcome prediction was assessed by different treatment inclusion criteria as the study's final phase. The MLP classifier had the highest performance among ConQuaFea models (AUC, SEN, SPE = 0.80, 0.85, 0.76). RF achieved the best performance in terms of AUC, SEN, and SPE with values of 0.65, 0.62, and 0.68, respectively, among radiomic models. GB and RF approaches achieved the best AUC, SEN, and SPE values of 0.78, 0.92, and 0.63 and 0.74, 0.93, and 0.56, respectively, among the combined models. A promising outcome was obtained when using radiomic and ConQuaFea from GSPECT MPI to detect left ventricular contractile patterns by machine learning.

Overview of the HECKTOR Challenge at MICCAI 2022: Automatic Head and Neck Tumor Segmentation and Outcome Prediction in PET/CT

This paper presents an overview of the third edition of the HEad and neCK TumOR segmentation and outcome prediction (HECKTOR) challenge, organized as a satellite event of the 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2022. The challenge comprises two tasks related to the automatic analysis of FDG-PET/CT images for patients with Head and Neck cancer (H&N), focusing on the oropharynx region. Task 1 is the fully automatic segmentation of H&N primary Gross Tumor Volume (GTVp) and metastatic lymph nodes (GTVn) from FDG-PET/CT images. Task 2 is the fully automatic prediction of Recurrence-Free Survival (RFS) from the same FDG-PET/CT and clinical data. The data were collected from nine centers for a total of 883 cases consisting of FDG-PET/CT images and clinical information, split into 524 training and 359 test cases. The best methods obtained an aggregated Dice Similarity Coefficient (DSCagg) of 0.788 in Task 1, and a Concordance index (C-index) of 0.682 in Task 2.

Artificial intelligence-based analysis of whole-body bone scintigraphy: The quest for the optimal deep learning algorithm and comparison with human observer performance

PURPOSE

Whole-body bone scintigraphy (WBS) is one of the most widely used modalities in diagnosing malignant bone diseases during the early stages. However, the procedure is time-consuming and requires vigour and experience. Moreover, interpretation of WBS scans in the early stages of the disorders might be challenging because the patterns often reflect normal appearance that is prone to subjective interpretation. To simplify the gruelling, subjective, and prone-to-error task of interpreting WBS scans, we developed deep learning (DL) models to automate two major analyses, namely (i) classification of scans into normal and abnormal and (ii) discrimination between malignant and non-neoplastic bone diseases, and compared their performance with human observers.

MATERIALS AND METHODS

After applying our exclusion criteria on 7188 patients from three different centers, 3772 and 2248 patients were enrolled for the first and second analyses, respectively. Data were split into two parts, including training and testing, while a fraction of training data were considered for validation. Ten different CNN models were applied to single- and dual-view input (posterior and anterior views) modes to find the optimal model for each analysis. In addition, three different methods, including squeeze-and-excitation (SE), spatial pyramid pooling (SPP), and attention-augmented (AA), were used to aggregate the features for dual-view input models. Model performance was reported through area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, and specificity and was compared with the DeLong test applied to ROC curves. The test dataset was evaluated by three nuclear medicine physicians (NMPs) with different levels of experience to compare the performance of AI and human observers.

RESULTS

DenseNet121_AA (DensNet121, with dual-view input aggregated by AA) and InceptionResNetV2_SPP achieved the highest performance (AUC = 0.72) for the first and second analyses, respectively. Moreover, on average, in the first analysis, Inception V3 and InceptionResNetV2 CNN models and dual-view input with AA aggregating method had superior performance. In addition, in the second analysis, DenseNet121 and InceptionResNetV2 as CNN methods and dual-view input with AA aggregating method achieved the best results. Conversely, the performance of AI models was significantly higher than human observers for the first analysis, whereas their performance was comparable in the second analysis, although the AI model assessed the scans in a drastically lower time.

CONCLUSION

Using the models designed in this study, a positive step can be taken toward improving and optimizing WBS interpretation. By training DL models with larger and more diverse cohorts, AI could potentially be used to assist physicians in the assessment of WBS images.

Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning

PURPOSE

Attenuation correction and scatter compensation (AC/SC) are two main steps toward quantitative PET imaging, which remain challenging in PET-only and PET/MRI systems. These can be effectively tackled via deep learning (DL) methods. However, trustworthy, and generalizable DL models commonly require well-curated, heterogeneous, and large datasets from multiple clinical centers. At the same time, owing to legal/ethical issues and privacy concerns, forming a large collective, centralized dataset poses significant challenges. In this work, we aimed to develop a DL-based model in a multicenter setting without direct sharing of data using federated learning (FL) for AC/SC of PET images.

METHODS

Non-attenuation/scatter corrected and CT-based attenuation/scatter corrected (CT-ASC) ¹⁸F-FDG PET images of 300 patients were enrolled in this study. The dataset consisted of 6 different centers, each with 50 patients, with scanner, image acquisition, and reconstruction protocols varying across the centers. CT-based ASC PET images served as the standard reference. All images were reviewed to include high-quality and artifact-free PET images. Both corrected and uncorrected PET images were converted to standardized uptake values (SUVs). We used a modified nested U-Net utilizing residual U-block in a U-shape architecture. We evaluated two FL models, namely sequential (FL-SQ) and parallel (FL-PL) and compared their performance with the baseline centralized (CZ) learning model wherein the data were pooled to one server, as well as center-based (CB) models where for each center the model was built and evaluated separately. Data from each center were divided to contribute to training (30 patients), validation (10 patients), and test sets (10 patients). Final evaluations and reports were performed on 60 patients (10 patients from each center).

RESULTS

In terms of percent SUV absolute relative error (ARE%), both FL-SQ (CI:12.21-14.81%) and FL-PL (CI:11.82-13.84%) models demonstrated excellent agreement with the centralized framework (CI:10.32-12.00%), while FL-based algorithms improved model performance by over 11% compared to CB training strategy (CI: 22.34-26.10%). Furthermore, the Mann-Whitney test between different strategies revealed no significant differences between CZ and FL-based algorithms (p-value > 0.05) in center-categorized mode. At the same time, a significant difference was observed between the different training approaches on the overall dataset (p-value < 0.05). In addition, voxel-wise comparison, with respect to reference CT-ASC, exhibited similar performance for images predicted by CZ (R² = 0.94), FL-SQ (R² = 0.93), and FL-PL (R² = 0.92), while CB model achieved a far lower coefficient of determination (R² = 0.74). Despite the strong correlations between CZ and FL-based methods compared to reference CT-ASC, a slight underestimation of predicted voxel values was observed.

CONCLUSION

Deep learning-based models provide promising results toward quantitative PET image reconstruction. Specifically, we developed two FL models and compared their performance with center-based and centralized models. The proposed FL-based models achieved higher performance compared to center-based models, comparable with centralized models. Our work provided strong empirical evidence that the FL framework can fully benefit from the generalizability and robustness of DL models used for AC/SC in PET, while obviating the need for the direct sharing of datasets between clinical imaging centers.

Patient-specific fetal radiation dosimetry for pregnant patients undergoing abdominal and pelvic CT imaging

BACKGROUND

Accurate estimation of fetal radiation dose is crucial for risk-benefit analysis of radiological imaging, while the radiation dosimetry studies based on individual pregnant patient are highly desired.

PURPOSE

To use Monte Carlo calculations for estimation of fetal radiation dose from abdominal and pelvic computed tomography (CT) examinations for a population of patients with a range of variations in patients' anatomy, abdominal circumference, gestational age (GA), fetal depth (FD), and fetal development.

METHODS

Forty-four patient-specific pregnant female models were constructed based on CT imaging data of pregnant patients, with gestational ages ranging from 8 to 35 weeks. The simulation of abdominal and pelvic helical CT examinations was performed on three validated commercial scanner systems to calculate organ-level fetal radiation dose.

RESULTS

The absorbed radiation dose to the fetus ranged between 0.97 and 2.24 mGy, with an average of 1.63 ± 0.33 mGy. The CTDI_vol -normalized fetal dose ranged between 0.56 and 1.30, with an average of 0.94 ± 0.25. The normalized fetal organ dose showed significant correlations with gestational age, maternal abdominal circumference (MAC), and fetal depth. The use of ATCM technique increased the fetal radiation dose in some patients.

CONCLUSION

A technique enabling the calculation of organ-level radiation dose to the fetus was developed from models of actual anatomy representing a range of gestational age, maternal size, and fetal position. The developed maternal and fetal models provide a basis for reliable and accurate radiation dose estimation to fetal organs.

Fully automated accurate patient positioning in computed tomography using anterior-posterior localizer images and a deep neural network: a dual-center study

OBJECTIVES

This study aimed to improve patient positioning accuracy by relying on a CT localizer and a deep neural network to optimize image quality and radiation dose.

METHODS

We included 5754 chest CT axial and anterior-posterior (AP) images from two different centers, C1 and C2. After pre-processing, images were split into training (80%) and test (20%) datasets. A deep neural network was trained to generate 3D axial images from the AP localizer. The geometric centerlines of patient bodies were indicated by creating a bounding box on the predicted images. The distance between the body centerline, estimated by the deep learning model and ground truth (BCAP), was compared with patient mis-centering during manual positioning (BCMP). We evaluated the performance of our model in terms of distance between the lung centerline estimated by the deep learning model and the ground truth (LCAP).

RESULTS

The error in terms of BCAP was - 0.75 ± 7.73 mm and 2.06 ± 10.61 mm for C1 and C2, respectively. This error was significantly lower than BCMP, which achieved an error of 9.35 ± 14.94 and 13.98 ± 14.5 mm for C1 and C2, respectively. The absolute BCAP was 5.7 ± 5.26 and 8.26 ± 6.96 mm for C1 and C2, respectively. The LCAP metric was 1.56 ± 10.8 and -0.27 ± 16.29 mm for C1 and C2, respectively. The error in terms of BCAP and LCAP was higher for larger patients (p value < 0.01).

CONCLUSION

The accuracy of the proposed method was comparable to available alternative methods, carrying the advantage of being free from errors related to objects blocking the camera visibility.

KEY POINTS

• Patient mis-centering in the anterior-posterior direction (AP) is a common problem in clinical practice which can degrade image quality and increase patient radiation dose. • We proposed a deep neural network for automatic patient positioning using only the CT image localizer, achieving a performance comparable to alternative techniques, such as the external 3D visual camera. • The advantage of the proposed method is that it is free from errors related to objects blocking the camera visibility and that it could be implemented on imaging consoles as a patient positioning support tool.

Triple-source saddle-curve cone-beam photon counting CT image reconstruction: A simulation study

PURPOSE

The most common detector material in the PC CT system, cannot achieve the best performance at a relatively higher photon flux rate. In the reconstruction view, the most commonly used filtered back projection, is not able to provide sufficient reconstructed image quality in spectral computed tomography (CT). Developing a triple-source saddle-curve cone-beam photon counting CT image reconstruction method can improve the temporal resolution.

METHODS

Triple-source saddle-curve cone-beam trajectory was rearranged into four trajectory sets for simulation and reconstruction. Projection images in different energy bins were simulated by forward projection and photon counting CT respond model simulation. After simulation, the object was reconstructed using Katsevich's theory after photon counts correction using the pseudo inverse of photon counting CT response matrix. The material decomposition can be performed based on images in different energy bins.

RESULTS

Root mean square error (RMSE) and structural similarity index (SSIM) are calculated to quantify the image quality of reconstruction images. Compared with FDK images, the RMSE for the triple-source image was improved by 27%, 21%, 14%, 8%, and 6% for the reconstrued image of 20-33, 33-47, 47-58, 58-69, 69-80 keV energy bin. The SSIM was improved by 1.031%, 0.665%, 0.396%, 0.235%, 0.174% for corresponding energy bin. The decomposition image based on corrected images shows improved RMSE and SSIM, each by 33.861% and 0.345%. SSIM of corrected decomposition image of iodine reaches 99.415% of the original image.

CONCLUSIONS

A new Triple-source saddle-curve cone-beam PC CT image reconstruction method was developed in this work. The exact reconstruction of the triple-source saddle-curve improved both the image quality and temporal resolution.

Two-step machine learning to diagnose and predict involvement of lungs in COVID-19 and pneumonia using CT radiomics

OBJECTIVE

To develop a two-step machine learning (ML) based model to diagnose and predict involvement of lungs in COVID-19 and non COVID-19 pneumonia patients using CT chest radiomic features.

METHODS

Three hundred CT scans (3-classes: 100 COVID-19, 100 pneumonia, and 100 healthy subjects) were enrolled in this study. Diagnostic task included 3-class classification. Severity prediction score for COVID-19 and pneumonia was considered as mild (0-25%), moderate (26-50%), and severe (>50%). Whole lungs were segmented utilizing deep learning-based segmentation. Altogether, 107 features including shape, first-order histogram, second and high order texture features were extracted. Pearson correlation coefficient (PCC≥90%) followed by different features selection algorithms were employed. ML-based supervised algorithms (Naïve Bays, Support Vector Machine, Bagging, Random Forest, K-nearest neighbors, Decision Tree and Ensemble Meta voting) were utilized. The optimal model was selected based on precision, recall and area-under-curve (AUC) by randomizing the training/validation, followed by testing using the test set.

RESULTS

Nine pertinent features (2 shape, 1 first-order, and 6 second-order) were obtained after features selection for both phases. In diagnostic task, the performance of 3-class classification using Random Forest was 0.909±0.026, 0.907±0.056, 0.902±0.044, 0.939±0.031, and 0.982±0.010 for precision, recall, F1-score, accuracy, and AUC, respectively. The severity prediction task using Random Forest achieved 0.868±0.123 precision, 0.865±0.121 recall, 0.853±0.139 F1-score, 0.934±0.024 accuracy, and 0.969±0.022 AUC.

CONCLUSION

The two-phase ML-based model accurately classified COVID-19 and pneumonia patients using CT radiomics, and adequately predicted severity of lungs involvement. This 2-steps model showed great potential in assessing COVID-19 CT images towards improved management of patients.