Models using comprehensive, lesion-level, longitudinal [68Ga]Ga-DOTA-TATE PET-derived features lead to superior outcome prediction in neuroendocrine tumor patients treated with [177Lu]Lu-DOTA-TATE.

PURPOSE: Somatostatin receptor (SSTR) imaging features are predictive of treatment outcome for neuroendocrine tumor (NET) patients receiving peptide receptor radionuclide therapy (PRRT). However, comprehensive (all metastatic lesions), longitudinal (temporal variation), and lesion-level measured features have never been explored. Such features allow for capturing the heterogeneity in disease response to treatment. Furthermore, models combining these features are lacking. In this work we evaluated the predictive power of comprehensive, longitudinal, lesion-level 68GA-SSTR-PET features combined with a multivariate linear regression (MLR) model. METHODS: This retrospective study enrolled NET patients treated with [177Lu]Lu-DOTA-TATE and imaged with [68Ga]Ga-DOTA-TATE at baseline and post-therapy. All lesions were segmented, anatomically labeled, and longitudinally matched. Lesion-level uptake and variation in uptake were measured. Patient-level features were engineered and selected for modeling of progression-free survival (PFS). The model was validated via concordance index, patient classification (ROC analysis), and survival analysis (Kaplan-Meier and Cox proportional hazards). The MLR was benchmarked against single feature predictions. RESULTS: Thirty-six NET patients were enrolled and stratified into poor and good responders (PFS ≥ 25 months). Four patient-level features were selected, the MLR concordance index was 0.826, and the AUC was 0.88 (0.85 specificity, 0.81 sensitivity). Survival analysis led to significant patient stratification (p<.001) and hazard ratio (3⨯10-5). Lastly, in a benchmark study, the MLR modeling approach outperformed all the single feature predictors. CONCLUSION: Comprehensive, lesion-level, longitudinal 68GA-SSTR-PET analysis, combined with MLR modeling, leads to excellent predictions of PRRT outcome in NET patients, outperforming non-comprehensive, patient-level, and single time-point feature predictions. MESSAGE: Neuroendocrine tumor, peptide receptor radionuclide therapy, Somatostatin Receptor Imaging, Outcome Prediction, Treatment Response Assessment.

Quantitative evaluation of lesion response heterogeneity for superior prognostication of clinical outcome.

PURPOSE: Standardized reporting of treatment response in oncology patients has traditionally relied on methods like RECIST, PERCIST and Deauville score. These endpoints assess only a few lesions, potentially overlooking the response heterogeneity of all disease. This study hypothesizes that comprehensive spatial-temporal evaluation of all individual lesions is necessary for superior prognostication of clinical outcome. METHODS: [18F]FDG PET/CT scans from 241 patients (127 diffuse large B-cell lymphoma (DLBCL) and 114 non-small cell lung cancer (NSCLC)) were retrospectively obtained at baseline and either during chemotherapy or post-chemoradiotherapy. An automated TRAQinform IQ software (AIQ Solutions) analyzed the images, performing quantification of change in regions of interest suspicious of cancer (lesion-ROI). Multivariable Cox proportional hazards (CoxPH) models were trained to predict overall survival (OS) with varied sets of quantitative features and lesion-ROI, compared by bootstrapping with C-index and t-tests. The best-fit model was compared to automated versions of previously established methods like RECIST, PERCIST and Deauville score. RESULTS: Multivariable CoxPH models demonstrated superior prognostic power when trained with features quantifying response heterogeneity in all individual lesion-ROI in DLBCL (C-index = 0.84, p < 0.001) and NSCLC (C-index = 0.71, p < 0.001). Prognostic power significantly deteriorated (p < 0.001) when using subsets of lesion-ROI (C-index = 0.78 and 0.67 for DLBCL and NSCLC, respectively) or excluding response heterogeneity (C-index = 0.67 and 0.70). RECIST, PERCIST, and Deauville score could not significantly associate with OS (C-index < 0.65 and p > 0.1), performing significantly worse than the multivariable models (p < 0.001). CONCLUSIONS: Quantitative evaluation of response heterogeneity of all individual lesions is necessary for the superior prognostication of clinical outcome.

Quantitative imaging biomarkers of immune-related adverse events in immune-checkpoint blockade-treated metastatic melanoma patients: a pilot study.

PURPOSE: To develop quantitative molecular imaging biomarkers of immune-related adverse event (irAE) development in malignant melanoma (MM) patients receiving immune-checkpoint inhibitors (ICI) imaged with 18F-FDG PET/CT. METHODS: 18F-FDG PET/CT images of 58 MM patients treated with anti-PD-1 or anti-CTLA-4 ICI were retrospectively analyzed for indication of irAE. Three target organs, most commonly affected by irAE, were considered: bowel, lung, and thyroid. Patient charts were reviewed to identify which patients experienced irAE, irAE grade, and time to irAE diagnosis. Target organs were segmented using a convolutional neural network (CNN), and novel quantitative imaging biomarkers - SUV percentiles (SUVX%) of 18F-FDG uptake within the target organs - were correlated with the clinical irAE status. Area under the receiver-operating characteristic curve (AUROC) was used to quantify irAE detection performance. Patients who did not experience irAE were used to establish normal ranges for target organ 18F-FDG uptake. RESULTS: A total of 31% (18/58) patients experienced irAE in the three target organs: bowel (n=6), lung (n=5), and thyroid (n=9). Optimal percentiles for identifying irAE were bowel (SUV95%, AUROC=0.79), lung (SUV95%, AUROC=0.98), and thyroid (SUV75%, AUROC=0.88). Optimal cut-offs for irAE detection were bowel (SUV95%>2.7 g/mL), lung (SUV95%>1.7 g/mL), and thyroid (SUV75%>2.1 g/mL). Normal ranges (95% confidence interval) for the SUV percentiles in patients without irAE were bowel [1.74, 2.86 g/mL], lung [0.73, 1.46 g/mL], and thyroid [0.86, 1.99 g/mL]. CONCLUSIONS: Increased 18F-FDG uptake within irAE-affected organs provides predictive information about the development of irAE in MM patients receiving ICI and represents a potential quantitative imaging biomarker for irAE. Some irAE can be detected on 18F-FDG PET/CT well before clinical symptoms appear.

Spatial process decomposition for quantitative imaging biomarkers using multiple images of varying shapes.

Quantitative imaging biomarkers (QIB) are extracted from medical images in radiomics for a variety of purposes including noninvasive disease detection, cancer monitoring, and precision medicine. The existing methods for QIB extraction tend to be ad hoc and not reproducible. In this article, a general and flexible statistical approach is proposed for handling up to three-dimensional medical images and reasonably capturing features with respect to specific spatial patterns. In particular, a model-based spatial process decomposition is developed where the random weights are unique to individual patients for component functions common across patients. Model fitting and selection are based on maximum likelihood, while feature extractions are via optimal prediction of the underlying true image. Simulation studies are conducted to investigate the properties of the proposed methodology. For illustration, a cancer image data set is analyzed and QIBs are extracted in association with a clinical endpoint.

Executive summary of AAPM Report Task Group 113: Guidance for the physics aspects of clinical trials.

The charge of AAPM Task Group 113 is to provide guidance for the physics aspects of clinical trials to minimize variability in planning and dose delivery for external beam trials involving photons and electrons. Several studies have demonstrated the importance of protocol compliance on patient outcome. Minimizing variability for treatments at different centers improves the quality and efficiency of clinical trials. Attention is focused on areas where variability can be minimized through standardization of protocols and processes through all aspects of clinical trials. Recommendations are presented for clinical trial designers, physicists supporting clinical trials at their individual clinics, quality assurance centers, and manufacturers.

A randomized trial of bevacizumab for newly diagnosed glioblastoma.

BACKGROUND: Concurrent treatment with temozolomide and radiotherapy followed by maintenance temozolomide is the standard of care for patients with newly diagnosed glioblastoma. Bevacizumab, a humanized monoclonal antibody against vascular endothelial growth factor A, is currently approved for recurrent glioblastoma. Whether the addition of bevacizumab would improve survival among patients with newly diagnosed glioblastoma is not known. METHODS: In this randomized, double-blind, placebo-controlled trial, we treated adults who had centrally confirmed glioblastoma with radiotherapy (60 Gy) and daily temozolomide. Treatment with bevacizumab or placebo began during week 4 of radiotherapy and was continued for up to 12 cycles of maintenance chemotherapy. At disease progression, the assigned treatment was revealed, and bevacizumab therapy could be initiated or continued. The trial was designed to detect a 25% reduction in the risk of death and a 30% reduction in the risk of progression or death, the two coprimary end points, with the addition of bevacizumab. RESULTS: A total of 978 patients were registered, and 637 underwent randomization. There was no significant difference in the duration of overall survival between the bevacizumab group and the placebo group (median, 15.7 and 16.1 months, respectively; hazard ratio for death in the bevacizumab group, 1.13). Progression-free survival was longer in the bevacizumab group (10.7 months vs. 7.3 months; hazard ratio for progression or death, 0.79). There were modest increases in rates of hypertension, thromboembolic events, intestinal perforation, and neutropenia in the bevacizumab group. Over time, an increased symptom burden, a worse quality of life, and a decline in neurocognitive function were more frequent in the bevacizumab group. CONCLUSIONS: First-line use of bevacizumab did not improve overall survival in patients with newly diagnosed glioblastoma. Progression-free survival was prolonged but did not reach the prespecified improvement target. (Funded by the National Cancer Institute; ClinicalTrials.gov number, NCT00884741.).

Uncertainty estimation and evaluation of deformation image registration based convolutional neural networks.

Objective.Fast and accurate deformable image registration (DIR), including DIR uncertainty estimation, is essential for safe and reliable clinical deployment. While recent deep learning models have shown promise in predicting DIR with its uncertainty, challenges persist in proper uncertainty evaluation and hyperparameter optimization for these methods. This work aims to develop and evaluate a model that can perform fast DIR and predict its uncertainty in seconds.Approach.This study introduces a novel probabilistic multi-resolution image registration model utilizing convolutional neural networks to estimate a multivariate normal distributed dense displacement field (DDF) in a multimodal image registration problem. To assess the quality of the DDF distribution predicted by the model, we propose a new metric based on the Kullback-Leibler divergence. The performance of our approach was evaluated against three other DIR algorithms (VoxelMorph, Monte Carlo dropout, and Monte Carlo B-spline) capable of predicting uncertainty. The evaluation of the models included not only the quality of the deformation but also the reliability of the estimated uncertainty. Our application investigated the registration of a treatment planning computed tomography (CT) to follow-up cone beam CT for daily adaptive radiotherapy.Main results.The hyperparameter tuning of the models showed a trade-off between the estimated uncertainty's reliability and the deformation's accuracy. In the optimal trade-off, our model excelled in contour propagation and uncertainty estimation (p <0.05) compared to existing uncertainty estimation models. We obtained an average dice similarity coefficient of 0.89 and a KL-divergence of 0.15.Significance.By addressing challenges in DIR uncertainty estimation and evaluation, our work showed that both the DIR and its uncertainty can be reliably predicted, paving the way for safe deployment in a clinical environment.

An automated methodology for whole-body, multimodality tracking of individual cancer lesions.

Objective. Manual analysis of individual cancer lesions to assess disease response is clinically impractical and requires automated lesion tracking methodologies. However, no methodology has been developed for whole-body individual lesion tracking, across an arbitrary number of scans, and acquired with various imaging modalities.Approach. This study introduces a lesion tracking methodology and benchmarked it using 2368Ga-DOTATATE PET/CT and PET/MR images of eight neuroendocrine tumor patients. The methodology consists of six steps: (1) alignment of multiple scans via image registration, (2) body-part labeling, (3) automatic lesion-wise dilation, (4) clustering of lesions based on local lesion shape metrics, (5) assignment of lesion tracks, and (6) output of a lesion graph. Registration performance was evaluated via landmark distance, lesion matching accuracy was evaluated between each image pair, and lesion tracking accuracy was evaluated via identical track ratio. Sensitivity studies were performed to evaluate the impact of lesion dilation (fixed versus automatic dilation), anatomic location, image modalities (inter- versus intra-modality), registration mode (direct versus indirect registration), and track size (number of time-points and lesions) on lesion matching and tracking performance.Main results. Manual contouring yielded 956 lesions, 1570 lesion-matching decisions, and 493 lesion tracks. The median residual registration error was 2.5 mm. The automatic lesion dilation led to 0.90 overall lesion matching accuracy, and an 88% identical track ratio. The methodology is robust regarding anatomic locations, image modalities, and registration modes. The number of scans had a moderate negative impact on the identical track ratio (94% for 2 scans, 91% for 3 scans, and 81% for 4 scans). The number of lesions substantially impacted the identical track ratio (93% for 2 nodes versus 54% for ≥5 nodes).Significance. The developed methodology resulted in high lesion-matching accuracy and enables automated lesion tracking in PET/CT and PET/MR.

Imaging of castration-resistant prostrate cancer: development of imaging response biomarkers.

PURPOSE OF REVIEW: The current standard for imaging castration-resistant prostate cancer (CRPC) focuses solely on detection. However, in order to assess treatment response, imaging must provide quantitative results that can be validated. RECENT FINDINGS: Bone scintigraphy remains the most commonly used imaging tool for CRPC in bone, but with limited quantification capabilities. Both PET and MRI provide quantitative measures that could be used to assess treatment response. Several PET tracers have been shown to be able to detect bone metastases, but more research regarding their use for treatment response assessment is necessary. Similarly, research has shown that diffusion-weighted and dynamic contrast-enhanced MRI can detect metastases, with some studies suggesting that they may be suitable for assessing treatment response. SUMMARY: Recent research has shown that many imaging techniques are able to successfully detect metastases in CRPC patients as well as or better than standard imaging. These imaging methods can also be applied to treatment response assessment; however, more research must be done to validate the quantitative measures before these techniques can be used clinically for assessing patients.

Quantification of radiation-induced lung damage with CT scans: the possible benefit for radiogenomics.

BACKGROUND: Radiation-induced lung damage (RILD) is an important problem. Although physical parameters such as the mean lung dose are used in clinical practice, they are not suited for individualised radiotherapy. Objective, quantitative measurements of RILD on a continuous instead of on an ordinal, semi-quantitative, semi-subjective scale, are needed. METHODS: Hounsfield unit (HU) changes before versus three months post-radiotherapy were correlated per voxel with the radiotherapy dose in 95 lung cancer patients. Deformable registration was used to register pre- and post-CT scans and the density increase was quantified for various dose bins. The dose-response curve for increased HU was quantified using the slope of a linear regression (HU/Gy). The end-point for the toxicity analysis was dyspnoea ≥ grade 2. RESULTS: Radiation dose was linearly correlated with the change in HU (mean R(2) = 0.74 ± 0.28). No differences in HU/Gy between groups treated with stereotactic radiotherapy, conventional radiotherapy alone, sequential or concurrent chemo- radiotherapy were observed. In the whole patient group, 33/95 (34.7%) had dyspnoea ≥ G2. Of the 48 patients with a HU/Gy below the median, 16 (33.3%) developed dyspnoea ≥ G2, while in the 47 patients with a HU/Gy above the median, 17 (36.1%) had dyspnoea ≥ G2 (not significant). Individual patients showed a nearly 21-fold difference in radiosensitivity, with HU/Gy ranging from 0 to 10 HU/Gy. CONCLUSIONS: HU changes identify objectively the whole range of individual radiosensitivity on a continuous, quantitative scale. CT density changes may allow more robust and accurate radiogenomics studies.

Radiotherapy for head and neck tumours in 2012 and beyond: conformal, tailored, and adaptive?

Intensity-modulated radiation therapy (IMRT) is a conformal irradiation technique that enables steep dose gradients. In head and neck tumours this approach spares parotid-gland function without compromise to treatment efficacy. Anatomical and molecular imaging modalities may be used to tailor treatment by enabling proper selection and delineation of target volumes and organs at risk, which in turn lead to dose prescriptions that take into account the underlying tumour biology (eg, human papillomavirus status). Therefore, adaptations can be made throughout the course of radiotherapy, as required. Planned dose increases to parts of the target volumes may also be used to match the radiosensitivity of tumours (so-called dose-painting), assessed by molecular imaging. For swift implementation of tailored and adaptive IMRT, tools and procedures, such as accurate image acquisition and reconstruction, automatic segmentation of target volumes and organs at risk, non-rigid image and dose registration, and dose summation methods, need to be developed and properly validated.

Comparison of automated full-body bone metastases delineation methods and their corresponding prognostic power.

Objective.Manual disease delineation in full-body imaging of patients with multiple metastases is often impractical due to high disease burden. However, this is a clinically relevant task as quantitative image techniques assessing individual metastases, while limited, have been shown to be predictive of treatment outcome. The goal of this work was to evaluate the efficacy of deep learning-based methods for full-body delineation of skeletal metastases and to compare their performance to existing methods in terms of disease delineation accuracy and prognostic power.Approach.1833 suspicious lesions on 3718F-NaF PET/CT scans of patients with metastatic castration-resistant prostate cancer (mCRPC) were contoured and classified as malignant, equivocal, or benign by a nuclear medicine physician. Two convolutional neural network (CNN) architectures (DeepMedic and nnUNet)were trained to delineate malignant disease regions with and without three-model ensembling. Malignant disease contours using previously established methods were obtained. The performance of each method was assessed in terms of four different tasks: (1) detection, (2) segmentation, (3) PET SUV metric correlations with physician-based data, and (4) prognostic power of progression-free survival.Main Results.The nnUnet three-model ensemble achieved superior detection performance with a mean (+/- standard deviation) sensitivity of 82.9±ccc 0.1% at the selected operating point. The nnUnet single and three-model ensemble achieved comparable segmentation performance with a mean Dice coefficient of 0.80±0.12 and 0.79±0.12, respectively, both outperforming other methods. The nnUNet ensemble achieved comparable or superior SUV metric correlation performance to gold-standard data. Despite superior disease delineation performance, the nnUNet methods did not display superior prognostic power over other methods.Significance.This work showed that CNN-based (nnUNet) methods are superior to the non-CNN methods for mCRPC disease delineation in full-body18F-NaF PET/CT. The CNN-based methods, however, do not hold greater prognostic power for predicting clinical outcome. This merits more investigation on the optimal selection of delineation methods for specific clinical tasks.

Performance of an automated registration-based method for longitudinal lesion matching and comparison to inter-reader variability.

Objective.Patients with metastatic disease are followed throughout treatment with medical imaging, and accurately assessing changes of individual lesions is critical to properly inform clinical decisions. The goal of this work was to assess the performance of an automated lesion-matching algorithm in comparison to inter-reader variability (IRV) of matching lesions between scans of metastatic cancer patients.Approach.Forty pairs of longitudinal PET/CT and CT scans were collected and organized into four cohorts: lung cancers, head and neck cancers, lymphomas, and advanced cancers. Cases were also divided by cancer burden: low-burden (<10 lesions), intermediate-burden (10-29), and high-burden (30+). Two nuclear medicine physicians conducted independent reviews of each scan-pair and manually matched lesions. Matching differences between readers were assessed to quantify the IRV of lesion matching. The two readers met to form a consensus, which was considered a gold standard and compared against the output of an automated lesion-matching algorithm. IRV and performance of the automated method were quantified using precision, recall, F1-score, and the number of differences.Main results.The performance of the automated method did not differ significantly from IRV for any metric in any cohort (p> 0.05, Wilcoxon paired test). In high-burden cases, the F1-score (median [range]) was 0.89 [0.63, 1.00] between the automated method and reader consensus and 0.93 [0.72, 1.00] between readers. In low-burden cases, F1-scores were 1.00 [0.40, 1.00] and 1.00 [0.40, 1.00], for the automated method and IRV, respectively. Automated matching was significantly more efficient than either reader (p< 0.001). In high-burden cases, median matching time for the readers was 60 and 30 min, respectively, while automated matching took a median of 3.9 minSignificance.The automated lesion-matching algorithm was successful in performing lesion matching, meeting the benchmark of IRV. Automated lesion matching can significantly expedite and improve the consistency of longitudinal lesion-matching.

Prospective inter- and intra-tracer repeatability analysis of radiomics features in [68Ga]Ga-PSMA-11 and [18F]F-PSMA-1007 PET scans in metastatic prostate cancer.

OBJECTIVE: This study aimed to quantify both the intra- and intertracer repeatability of lesion-level radiomics features in [68Ga]Ga-prostate-specific membrane antigen (PSMA)-11 and [18F]F-PSMA-1007 positron emission tomography (PET) scans. METHODS: Eighteen patients with metastatic prostate cancer (mPCa) were prospectively recruited for the study and randomised to one of three test-retest groups: (i) intratracer [68Ga]Ga-PSMA-11 PET, (ii) intratracer [18F]F-PSMA-1007 PET or (iii) intertracer between [68Ga]Ga-PSMA-11 and [18F]F-PSMA-1007 PET. Four conventional PET metrics (standardised uptake value (SUV)max, SUVmean, SUVtotal and volume) and 107 radiomics features were extracted from 75 lesions and assessed using the repeatability coefficient (RC) and the ICC. Radiomic feature repeatability was also quantified after the application of 16 filters to the PET image. RESULTS: Test-retest scans were taken a median of 5 days apart (range: 2-7 days). SUVmean demonstrated the lowest RC limits of the conventional features, with RCs of 7.9%, 14.2% and 24.7% for the [68Ga]Ga-PSMA-11 PET, [18F]F-PSMA-1007 PET, and intertracer groups, respectively. 69%, 66% and 9% of all radiomics features had good or excellent ICC values (ICC ≥ 0.75) for the same groups. Feature repeatability therefore diminished considerably for the intertracer group relative to intratracer groups. CONCLUSION: In this study, robust biomarkers for each tracer group that can be used in subsequent clinical studies were identified. Overall, the repeatability of conventional and radiomic features were found to be substantially lower for the intertracer group relative to both intratracer groups, suggesting that assessing patient response quantitatively should be done using the same radiotracer where possible. ADVANCES IN KNOWLEDGE: Intertracer biomarker repeatability limits are significantly larger than intratracer limits.

Uncertainty estimation for deep learning-based pectoral muscle segmentation via Monte Carlo dropout.

Objective. Deep Learning models are often susceptible to failures after deployment. Knowing when your model is producing inadequate predictions is crucial. In this work, we investigate the utility of Monte Carlo (MC) dropout and the efficacy of the proposed uncertainty metric (UM) for flagging of unacceptable pectoral muscle segmentations in mammograms.Approach. Segmentation of pectoral muscle was performed with modified ResNet18 convolutional neural network. MC dropout layers were kept unlocked at inference time. For each mammogram, 50 pectoral muscle segmentations were generated. The mean was used to produce the final segmentation and the standard deviation was applied for the estimation of uncertainty. From each pectoral muscle uncertainty map, the overall UM was calculated. To validate the UM, a correlation between the dice similarity coefficient (DSC) and UM was used. The UM was first validated in a training set (200 mammograms) and finally tested in an independent dataset (300 mammograms). ROC-AUC analysis was performed to test the discriminatory power of the proposed UM for flagging unacceptable segmentations.Main results. The introduction of dropout layers in the model improved segmentation performance (DSC = 0.95 ± 0.07 versus DSC = 0.93 ± 0.10). Strong anti-correlation (r= -0.76,p< 0.001) between the proposed UM and DSC was observed. A high AUC of 0.98 (97% specificity at 100% sensitivity) was obtained for the discrimination of unacceptable segmentations. Qualitative inspection by the radiologist revealed that images with high UM are difficult to segment.Significance. The use of MC dropout at inference time in combination with the proposed UM enables flagging of unacceptable pectoral muscle segmentations from mammograms with excellent discriminatory power.

Challenges and opportunities for implementing hypofractionated radiotherapy in Africa: lessons from the HypoAfrica clinical trial.

The rising cancer incidence and mortality in sub-Saharan Africa (SSA) warrants an increased focus on adopting or developing approaches that can significantly increase access to treatment in the region. One such approach recommended by the recent Lancet Oncology Commission for sub-Saharan Africa is hypofractionated radiotherapy (HFRT), which can substantially increase access to radiotherapy by reducing the overall duration of time (in days) each person spends being treated. Here we highlight challenges in adopting such an approach identified during the implementation of the HypoAfrica clinical trial. The HypoAfrica clinical trial is a longitudinal, multicentre study exploring the feasibility of applying HFRT for prostate cancer in SSA. This study has presented an opportunity for a pragmatic assessment of potential barriers and facilitators to adopting HFRT. Our results highlight three key challenges: quality assurance, study harmonisation and machine maintenance. We describe solutions employed to resolve these challenges and opportunities for longer term solutions that can facilitate scaling-up use of HFRT in SSA in clinical care and multicentre clinical trials. This report provides a valuable reference for the utilisation of radiotherapy approaches that increase access to treatment and the conduct of high-quality large-scale/multi-centre clinical trials involving radiotherapy. Trial registration: Not available yet.

Development of a deep learning network for Alzheimer's disease classification with evaluation of imaging modality and longitudinal data.

Objective. Neuroimaging uncovers important information about disease in the brain. Yet in Alzheimer's disease (AD), there remains a clear clinical need for reliable tools to extract diagnoses from neuroimages. Significant work has been done to develop deep learning (DL) networks using neuroimaging for AD diagnosis. However, no particular model has emerged as optimal. Due to a lack of direct comparisons and evaluations on independent data, there is no consensus on which modality is best for diagnostic models or whether longitudinal information enhances performance. The purpose of this work was (1) to develop a generalizable DL model to distinguish neuroimaging scans of AD patients from controls and (2) to evaluate the influence of imaging modality and longitudinal data on performance.Approach. We trained a 2-class convolutional neural network (CNN) with and without a cascaded recurrent neural network (RNN). We used datasets of 772 (NAD = 364,Ncontrol= 408) 3D18F-FDG PET scans and 780 (NAD = 280,Ncontrol= 500) T1-weighted volumetric-3D MR images (containing 131 and 144 patients with multiple timepoints) from the Alzheimer's Disease Neuroimaging Initiative, plus an independent set of 104 (NAD = 63,NNC = 41)18F-FDG PET scans (one per patient) for validation.Main Results. ROC analysis showed that PET-trained models outperformed MRI-trained, achieving maximum AUC with the CNN + RNN model of 0.93 ± 0.08, with accuracy 82.5 ± 8.9%. Adding longitudinal information offered significant improvement to performance on18F-FDG PET, but not on T1-MRI. CNN model validation with an independent18F-FDG PET dataset achieved AUC of 0.99. Layer-wise relevance propagation heatmaps added CNN interpretability.Significance. The development of a high-performing tool for AD diagnosis, with the direct evaluation of key influences, reveals the advantage of using18F-FDG PET and longitudinal data over MRI and single timepoint analysis. This has significant implications for the potential of neuroimaging for future research on AD diagnosis and clinical management of suspected AD patients.

Federated learning enables big data for rare cancer boundary detection.

Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing.

Interpretation and visualization techniques for deep learning models in medical imaging.

Deep learning (DL) approaches to medical image analysis tasks have recently become popular; however, they suffer from a lack of human interpretability critical for both increasing understanding of the methods' operation and enabling clinical translation. This review summarizes currently available methods for performing image model interpretation and critically evaluates published uses of these methods for medical imaging applications. We divide model interpretation in two categories: (1) understanding model structure and function and (2) understanding model output. Understanding model structure and function summarizes ways to inspect the learned features of the model and how those features act on an image. We discuss techniques for reducing the dimensionality of high-dimensional data and cover autoencoders, both of which can also be leveraged for model interpretation. Understanding model output covers attribution-based methods, such as saliency maps and class activation maps, which produce heatmaps describing the importance of different parts of an image to the model prediction. We describe the mathematics behind these methods, give examples of their use in medical imaging, and compare them against one another. We summarize several published toolkits for model interpretation specific to medical imaging applications, cover limitations of current model interpretation methods, provide recommendations for DL practitioners looking to incorporate model interpretation into their task, and offer general discussion on the importance of model interpretation in medical imaging contexts.