Search

This study evaluates metrics for tasks such as classification, regression, clustering, correlation analysis, statistical tests, segmentation, and image-to-image (I2I) translation. Metrics were compared across Python libraries, R packages, and Matlab functions to assess their consistency and highlight discrepancies. The findings underscore the need for a unified roadmap to standardize metrics, ensuring reliable and reproducible ML evaluations across platforms. This study examined a wide range of evaluation metrics across various tasks and found only some to be consistent across platforms, such as (i) Accuracy, Balanced Accuracy, Cohens Kappa, F-beta Score, MCC, Geometric Mean, AUC, and Log Loss in binary classification; (ii) Accuracy, Cohens Kappa, and F-beta Score in multi-class classification; (iii) MAE, MSE, RMSE, MAPE, Explained Variance, Median AE, MSLE, and Huber in regression; (iv) Davies-Bouldin Index and Calinski-Harabasz Index in clustering; (v) Pearson, Spearman, Kendall's Tau, Mutual Information, Distance Correlation, Percbend, Shepherd, and Partial Correlation in correlation analysis; (vi) Paired t-test, Chi-Square Test, ANOVA, Kruskal-Wallis Test, Shapiro-Wilk Test, Welchs t-test, and Bartlett's test in statistical tests; (vii) Accuracy, Precision, and Recall in 2D segmentation; (viii) Accuracy in 3D segmentation; (ix) MAE, MSE, RMSE, and R-Squared in 2D-I2I translation; and (x) MAE, MSE, and RMSE in 3D-I2I translation. Given observation of discrepancies in a number of metrics (e.g. precision, recall and F1 score in binary classification, WCSS in clustering, multiple statistical tests, and IoU in segmentation, amongst multiple metrics), this study concludes that ML evaluation metrics require standardization and recommends that future research use consistent metrics for different tasks to effectively compare ML techniques and solutions., Comment: This paper is 12 pages with 1 table and 10 figures

Radiopharmaceutical therapies (RPTs) present a major opportunity to improve cancer therapy. Although many current RPTs use the same injected activity for all patients, there is interest in using absorbed dose measurements to enable personalized prescriptions. Image-based absorbed dose calculations incur uncertainties from calibration factors, partial volume effects and segmentation methods. While previously published dose estimation protocols incorporate these uncertainties, they do not account for uncertainty originating from the reconstruction process itself with the propagation of Poisson noise from projection data. This effect should be accounted for to adequately estimate the total uncertainty in absorbed dose estimates. This paper proposes a computationally practical algorithm that propagates uncertainty from projection data through clinical reconstruction algorithms to obtain uncertainties on the total measured counts within volumes of interest (VOIs). The algorithm is first validated on ${}^{177}$Lu and ${}^{225}$Ac phantom data by comparing estimated uncertainties from individual SPECT acquisitions to empirical estimates obtained from multiple acquisitions. It is then applied to (i) Monte Carlo and (ii) multi-time point ${}^{177}$Lu-DOTATATE and ${}^{225}$Ac-PSMA-617 patient data for time integrated activity (TIA) uncertainty estimation. The outcomes of this work are two-fold: (i) the proposed uncertainty algorithm is validated, and (ii) a blueprint is established for how the algorithm can be inform dosimetry protocols via TIA uncertainty estimation. The proposed algorithm is made publicly available in the open-source image reconstruction library PyTomography.

Modeling of the collimator-detector response (CDR) in SPECT reconstruction enables improved resolution and more accurate quantitation, especially for higher energy imaging (e.g.Lu-177 and Ac-225). Such modeling, however, can pose a significant computational bottleneck when there are substantial components of septal penetration and scatter in the acquired data, since a direct convolution-based approach requires large 2D kernels. The present work presents an alternative method for fast and accurate CDR compensation using a linear operator built from 1D convolutions and rotations (1D-R). To enable open-source development and use of these models in image reconstruction, we release a SPECTPSFToolbox repository for the PyTomography project on GitHub.

Purpose: Artificial intelligence (AI) techniques have been extensively utilized for diagnosing and prognosis of several diseases in recent years. This study identifies, appraises and synthesizes published studies on the use of AI for the prognosis of COVID-19. Method: Electronic search was performed using Medline, Google Scholar, Scopus, Embase, Cochrane and ProQuest. Studies that examined machine learning or deep learning methods to determine the prognosis of COVID-19 using CT or chest X-ray images were included. Polled sensitivity, specificity area under the curve and diagnostic odds ratio were calculated. Result: A total of 36 articles were included; various prognosis-related issues, including disease severity, mechanical ventilation or admission to the intensive care unit and mortality, were investigated. Several AI models and architectures were employed, such as the Siamense model, support vector machine, Random Forest , eXtreme Gradient Boosting, and convolutional neural networks. The models achieved 71%, 88% and 67% sensitivity for mortality, severity assessment and need for ventilation, respectively. The specificity of 69%, 89% and 89% were reported for the aforementioned variables. Conclusion: Based on the included articles, machine learning and deep learning methods used for the prognosis of COVID-19 patients using radiomic features from CT or CXR images can help clinicians manage patients and allocate resources more effectively. These studies also demonstrate that combining patient demographic, clinical data, laboratory tests and radiomic features improves model performances.

Prostate specific membrane antigen (PSMA) positron emission tomography/computed tomography (PET/CT) imaging provides a tremendously exciting frontier in visualization of prostate cancer (PCa) metastatic lesions. However, accurate segmentation of metastatic lesions is challenging due to low signal-to-noise ratios and variable sizes, shapes, and locations of the lesions. This study proposes a novel approach for automated segmentation of metastatic lesions in PSMA PET/CT 3D volumetric images using 2D denoising diffusion probabilistic models (DDPMs). Instead of 2D trans-axial slices or 3D volumes, the proposed approach segments the lesions on generated multi-angle maximum intensity projections (MA-MIPs) of the PSMA PET images, then obtains the final 3D segmentation masks from 3D ordered subset expectation maximization (OSEM) reconstruction of 2D MA-MIPs segmentations. Our proposed method achieved superior performance compared to state-of-the-art 3D segmentation approaches in terms of accuracy and robustness in detecting and segmenting small metastatic PCa lesions. The proposed method has significant potential as a tool for quantitative analysis of metastatic burden in PCa patients., Comment: 11 pages, 2 figures, accepted in the DGM4MICCAI workshop, MICCAI, 2024

The objective of this study was to develop an automated pipeline that enhances thyroid disease classification using thyroid scintigraphy images, aiming to decrease assessment time and increase diagnostic accuracy. Anterior thyroid scintigraphy images from 2,643 patients were collected and categorized into diffuse goiter (DG), multinodal goiter (MNG), and thyroiditis (TH) based on clinical reports, and then segmented by an expert. A ResUNet model was trained to perform auto-segmentation. Radiomic features were extracted from both physician (scenario 1) and ResUNet segmentations (scenario 2), followed by omitting highly correlated features using Spearman's correlation, and feature selection using Recursive Feature Elimination (RFE) with XGBoost as the core. All models were trained under leave-one-center-out cross-validation (LOCOCV) scheme, where nine instances of algorithms were iteratively trained and validated on data from eight centers and tested on the ninth for both scenarios separately. Segmentation performance was assessed using the Dice similarity coefficient (DSC), while classification performance was assessed using metrics, such as precision, recall, F1-score, accuracy, area under the Receiver Operating Characteristic (ROC AUC), and area under the precision-recall curve (PRC AUC). ResUNet achieved DSC values of 0.84$\pm$0.03, 0.71$\pm$0.06, and 0.86$\pm$0.02 for MNG, TH, and DG, respectively. Classification in scenario 1 achieved an accuracy of 0.76$\pm$0.04 and a ROC AUC of 0.92$\pm$0.02 while in scenario 2, classification yielded an accuracy of 0.74$\pm$0.05 and a ROC AUC of 0.90$\pm$0.02. The automated pipeline demonstrated comparable performance to physician segmentations on several classification metrics across different classes, effectively reducing assessment time while maintaining high diagnostic accuracy. Code available at: https://github.com/ahxmeds/thyroidiomics.git., Comment: 7 pages, 4 figures, 2 tables

Optimal experimental design is a well studied field in applied science and engineering. Techniques for estimating such a design are commonly used within the framework of parameter estimation. Nonetheless, in recent years parameter estimation techniques are changing rapidly with the introduction of deep learning techniques to replace traditional estimation methods. This in turn requires the adaptation of optimal experimental design that is associated with these new techniques. In this paper we investigate a new experimental design methodology that uses deep learning. We show that the training of a network as a Likelihood Free Estimator can be used to significantly simplify the design process and circumvent the need for the computationally expensive bi-level optimization problem that is inherent in optimal experimental design for non-linear systems. Furthermore, deep design improves the quality of the recovery process for parameter estimation problems. As proof of concept we apply our methodology to two different systems of Ordinary Differential Equations.

The 2nd SNMMI Artificial Intelligence (AI) Summit, organized by the SNMMI AI Task Force, took place in Bethesda, MD, on February 29 - March 1, 2024. Bringing together various community members and stakeholders, and following up on a prior successful 2022 AI Summit, the summit theme was: AI in Action. Six key topics included (i) an overview of prior and ongoing efforts by the AI task force, (ii) emerging needs and tools for computational nuclear oncology, (iii) new frontiers in large language and generative models, (iv) defining the value proposition for the use of AI in nuclear medicine, (v) open science including efforts for data and model repositories, and (vi) issues of reimbursement and funding. The primary efforts, findings, challenges, and next steps are summarized in this manuscript.

Dynamic Positron Emission Tomography (dPET) imaging and Time-Activity Curve (TAC) analyses are essential for understanding and quantifying the biodistribution of radiopharmaceuticals over time and space. Traditional compartmental modeling, while foundational, commonly struggles to fully capture the complexities of biological systems, including non-linear dynamics and variability. This study introduces an innovative data-driven neural network-based framework, inspired by Reaction Diffusion systems, designed to address these limitations. Our approach, which adaptively fits TACs from dPET, enables the direct calibration of diffusion coefficients and reaction terms from observed data, offering significant improvements in predictive accuracy and robustness over traditional methods, especially in complex biological scenarios. By more accurately modeling the spatio-temporal dynamics of radiopharmaceuticals, our method advances modeling of pharmacokinetic and pharmacodynamic processes, enabling new possibilities in quantitative nuclear medicine.

Towards the need for automated and precise AI-based analysis of medical images, we present RT-utils, a specialized Python library tuned for the manipulation of radiotherapy (RT) structures stored in DICOM format. RT-utils excels in converting the polygon contours into binary masks, ensuring accuracy and efficiency. By converting DICOM RT structures into standardized formats such as NumPy arrays and SimpleITK Images, RT-utils optimizes inputs for computational solutions such as AI-based automated segmentation techniques or radiomics analysis. Since its inception in 2020, RT-utils has been used extensively with a focus on simplifying complex data processing tasks. RT-utils offers researchers a powerful solution to enhance workflows and drive significant advancements in medical imaging.

$^{18}$F-Fluoromisonidazole ($^{18}$F-FMISO) is a highly promising positron emission tomography radiopharmaceutical for identifying hypoxic regions in solid tumors. This research employs spatiotemporal multi-scale mathematical modeling to explore how different levels of angiogenesis influence the transport of radiopharmaceuticals within tumors. In this study, two tumor geometries with heterogeneous and uniform distributions of capillary networks were employed to incorporate varying degrees of microvascular density. The synthetic image of the heterogeneous and vascularized tumor was generated by simulating the angiogenesis process. The proposed multi-scale spatiotemporal model accounts for intricate physiological and biochemical factors within the tumor microenvironment, such as the transvascular transport of the radiopharmaceutical agent, its movement into the interstitial space by diffusion and convection mechanisms, and ultimately its uptake by tumor cells. Results showed that both quantitative and semi-quantitative metrics of $^{18}$F-FMISO uptake differ spatially and temporally at different stages during tumor growth. The presence of a high microvascular density in uniformly vascularized tumor increases cellular uptake, as it allows for more efficient release and rapid distribution of radiopharmaceutical molecules. This results in enhanced uptake compared to the heterogeneous vascularized tumor. In both heterogeneous and uniform distribution of microvessels in tumors, the diffusion transport mechanism has a more pronounced than convection. The findings of this study shed light on the transport phenomena behind $^{18}$F-FMISO radiopharmaceutical distribution and its delivery in the tumor microenvironment, aiding oncologists in their routine decision-making processes.

We introduce an innovative, simple, effective segmentation-free approach for outcome prediction in head \& neck cancer (HNC) patients. By harnessing deep learning-based feature extraction techniques and multi-angle maximum intensity projections (MA-MIPs) applied to Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) volumes, our proposed method eliminates the need for manual segmentations of regions-of-interest (ROIs) such as primary tumors and involved lymph nodes. Instead, a state-of-the-art object detection model is trained to perform automatic cropping of the head and neck region on the PET volumes. A pre-trained deep convolutional neural network backbone is then utilized to extract deep features from MA-MIPs obtained from 72 multi-angel axial rotations of the cropped PET volumes. These deep features extracted from multiple projection views of the PET volumes are then aggregated and fused, and employed to perform recurrence-free survival analysis on a cohort of 489 HNC patients. The proposed approach outperforms the best performing method on the target dataset for the task of recurrence-free survival analysis. By circumventing the manual delineation of the malignancies on the FDG PET-CT images, our approach eliminates the dependency on subjective interpretations and highly enhances the reproducibility of the proposed survival analysis method., Comment: 15 pages, 4 tables, 4 figures. Published in Cancers 2024, Volume 16, Issue 14, page 2538

Minimizing the need for pixel-level annotated data for training PET anomaly segmentation networks is crucial, particularly due to time and cost constraints related to expert annotations. Current un-/weakly-supervised anomaly detection methods rely on autoencoder or generative adversarial networks trained only on healthy data, although these are more challenging to train. In this work, we present a weakly supervised and Implicitly guided COuNterfactual diffusion model for Detecting Anomalies in PET images, branded as IgCONDA-PET. The training is conditioned on image class labels (healthy vs. unhealthy) along with implicit guidance to generate counterfactuals for an unhealthy image with anomalies. The counterfactual generation process synthesizes the healthy counterpart for a given unhealthy image, and the difference between the two facilitates the identification of anomaly locations. The code is available at: https://github.com/igcondapet/IgCONDA-PET.git, Comment: 12 pages, 6 figures, 1 table

For whole-body (WB) kinetic modeling based on a typical PET scanner, a multi-pass multi-bed scanning protocol is necessary given the limited axial field-of-view. Such a protocol introduces loss of early-dynamics in time-activity curves (TACs) and sparsity in TAC measurements, inducing uncertainty in parameter estimation when using least-squares estimation (LSE) (i.e., common standard) especially for kinetic microparameters. We present a method to reliably estimate microparameters, enabling accurate parametric imaging, on regular-axial field-of-view PET scanners. Our method, denoted parameter combination-driven estimation (PCDE), relies on generation of reference truth TAC database, and subsequently selecting, the best parameter combination as the one arriving at TAC with highest total similarity score (TSS), focusing on general image quality, overall visibility, and tumor detectability metrics. Our technique has two distinctive characteristics: 1) improved probability of having one-on-one mapping between early and late dynamics in TACs (the former missing from typical protocols), and 2) use of multiple aspects of TACs in selection of best fits. To evaluate our method against conventional LSE, we plotted tradeoff curves for noise and bias. In addition, the overall SNR and spatial noise were calculated and compared. Furthermore, CNR and TBR were also calculated. We also tested our proposed method on patient data (18F-DCFPyL PET scans) to further verify clinical applicability. Significantly improved general image quality performance was verified in microparametric images (e.g. noise-bias performance). The overall visibility and tumor detectability were also improved. Finally, for our patient studies, improved overall visibility and tumor detectability were demonstrated in micoparametric images, compared to use of conventional parameter estimation.

The effectiveness of Deep Neural Networks (DNNs) heavily relies on the abundance and accuracy of available training data. However, collecting and annotating data on a large scale is often both costly and time-intensive, particularly in medical cases where practitioners are already occupied with their duties. Moreover, ensuring that the model remains robust across various scenarios of image capture is crucial in medical domains, especially when dealing with ultrasound images that vary based on the settings of different devices and the manual operation of the transducer. To address this challenge, we introduce a novel pipeline called MEDDAP, which leverages Stable Diffusion (SD) models to augment existing small datasets by automatically generating new informative labeled samples. Pretrained checkpoints for SD are typically based on natural images, and training them for medical images requires significant GPU resources due to their heavy parameters. To overcome this challenge, we introduce USLoRA (Ultrasound Low-Rank Adaptation), a novel fine-tuning method tailored specifically for ultrasound applications. USLoRA allows for selective fine-tuning of weights within SD, requiring fewer than 0.1\% of parameters compared to fully fine-tuning only the UNet portion of SD. To enhance dataset diversity, we incorporate different adjectives into the generation process prompts, thereby desensitizing the classifiers to intensity changes across different images. This approach is inspired by clinicians' decision-making processes regarding breast tumors, where tumor shape often plays a more crucial role than intensity. In conclusion, our pipeline not only outperforms classifiers trained on the original dataset but also demonstrates superior performance when encountering unseen datasets. The source code is available at https://github.com/yasamin-med/MEDDAP., Comment: submitted to miccai 2024 submitted to miccai 2024 Submitted to MICCAI-2024

Automated slice classification is clinically relevant since it can be incorporated into medical image segmentation workflows as a preprocessing step that would flag slices with a higher probability of containing tumors, thereby directing physicians attention to the important slices. In this work, we train a ResNet-18 network to classify axial slices of lymphoma PET/CT images (collected from two institutions) depending on whether the slice intercepted a tumor (positive slice) in the 3D image or if the slice did not (negative slice). Various instances of the network were trained on 2D axial datasets created in different ways: (i) slice-level split and (ii) patient-level split; inputs of different types were used: (i) only PET slices and (ii) concatenated PET and CT slices; and different training strategies were employed: (i) center-aware (CAW) and (ii) center-agnostic (CAG). Model performances were compared using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC), and various binary classification metrics. We observe and describe a performance overestimation in the case of slice-level split as compared to the patient-level split training. The model trained using patient-level split data with the network input containing only PET slices in the CAG training regime was the best performing/generalizing model on a majority of metrics. Our models were additionally more closely compared using the sensitivity metric on the positive slices from their respective test sets., Comment: 10 pages, 6 figures, 2 tables

Accurate detection and segmentation of diffuse large B-cell lymphoma (DLBCL) from PET images has important implications for estimation of total metabolic tumor volume, radiomics analysis, surgical intervention and radiotherapy. Manual segmentation of tumors in whole-body PET images is time-consuming, labor-intensive and operator-dependent. In this work, we develop and validate a fast and efficient three-step cascaded deep learning model for automated detection and segmentation of DLBCL tumors from PET images. As compared to a single end-to-end network for segmentation of tumors in whole-body PET images, our three-step model is more effective (improves 3D Dice score from 58.9% to 78.1%) since each of its specialized modules, namely the slice classifier, the tumor detector and the tumor segmentor, can be trained independently to a high degree of skill to carry out a specific task, rather than a single network with suboptimal performance on overall segmentation., Comment: 8 pages, 3 figures, 3 tables

The purpose was to investigate the spatial heterogeneity of prostate-specific membrane antigen (PSMA) positron emission tomography (PET) uptake within parotid glands. We aim to quantify patterns in well-defined regions to facilitate further investigations. Furthermore, we investigate whether uptake is correlated with computed tomography (CT) texture features. Parotid glands from [18F]DCFPyL PSMA PET/CT images of 30 prostate cancer patients were analyzed. Thresholding was used to define high-uptake regions, and uptake statistics were computed within various divisions. Spearman's rank correlation coefficient was calculated between PSMA PET uptake and the Grey Level Run Length Matrix (GLRLM) using a long and short run length emphasis (GLRLML and GLRLMS) in subregions of parotid glands. PSMA PET uptake was significantly higher (p < 0.001) in lateral/posterior regions of the glands than anterior/medial regions. Maximum uptake was found in the lateral half of parotid glands in 50 out of 60 glands. The difference in SUV between parotid halves is greatest when parotids are divided by a plane separating the anterior/medial and posterior/lateral halves symmetrically. PSMA PET uptake was significantly correlated with CT GLRLML (p < 0.001), and anti-correlated with CT GLRLMS (p < 0.001). Uptake of PSMA PET is heterogeneous within parotid glands, with uptake biased towards lateral and posterior regions. Uptake patterns within parotid glands were found to be strongly correlated with CT texture features, suggesting the possible future use of CT texture features as a proxy for inferring PSMA PET uptake in salivary glands.

This study performs comprehensive evaluation of four neural network architectures (UNet, SegResNet, DynUNet, and SwinUNETR) for lymphoma lesion segmentation from PET/CT images. These networks were trained, validated, and tested on a diverse, multi-institutional dataset of 611 cases. Internal testing (88 cases; total metabolic tumor volume (TMTV) range [0.52, 2300] ml) showed SegResNet as the top performer with a median Dice similarity coefficient (DSC) of 0.76 and median false positive volume (FPV) of 4.55 ml; all networks had a median false negative volume (FNV) of 0 ml. On the unseen external test set (145 cases with TMTV range: [0.10, 2480] ml), SegResNet achieved the best median DSC of 0.68 and FPV of 21.46 ml, while UNet had the best FNV of 0.41 ml. We assessed reproducibility of six lesion measures, calculated their prediction errors, and examined DSC performance in relation to these lesion measures, offering insights into segmentation accuracy and clinical relevance. Additionally, we introduced three lesion detection criteria, addressing the clinical need for identifying lesions, counting them, and segmenting based on metabolic characteristics. We also performed expert intra-observer variability analysis revealing the challenges in segmenting ``easy'' vs. ``hard'' cases, to assist in the development of more resilient segmentation algorithms. Finally, we performed inter-observer agreement assessment underscoring the importance of a standardized ground truth segmentation protocol involving multiple expert annotators. Code is available at: https://github.com/microsoft/lymphoma-segmentation-dnn, Comment: 12 pages, 10 figures, 2 tables

Automated segmentation of cancerous lesions in PET/CT images is a vital initial task for quantitative analysis. However, it is often challenging to train deep learning-based segmentation methods to high degree of accuracy due to the diversity of lesions in terms of their shapes, sizes, and radiotracer uptake levels. These lesions can be found in various parts of the body, often close to healthy organs that also show significant uptake. Consequently, developing a comprehensive PET/CT lesion segmentation model is a demanding endeavor for routine quantitative image analysis. In this work, we train a 3D Residual UNet using Generalized Dice Focal Loss function on the AutoPET challenge 2023 training dataset. We develop our models in a 5-fold cross-validation setting and ensemble the five models via average and weighted-average ensembling. On the preliminary test phase, the average ensemble achieved a Dice similarity coefficient (DSC), false-positive volume (FPV) and false negative volume (FNV) of 0.5417, 0.8261 ml, and 0.2538 ml, respectively, while the weighted-average ensemble achieved 0.5417, 0.8186 ml, and 0.2538 ml, respectively. Our algorithm can be accessed via this link: https://github.com/ahxmeds/autosegnet., Comment: AutoPET-II challenge (2023)

Xerostomia and radiation-induced salivary gland dysfunction remain a common side effect for head-and-neck radiotherapy patients, and attempts have been made to quantify the heterogeneous dose response within parotid glands. Here several models of parotid gland subregional importance are compared with prostate specific membrane antigen (PSMA) positron emission tomography (PET) uptake. PSMA ligands show high concentrations in salivary glands, whose uptake has been previously found to relate to gland functionality. We develop a predictive model for relative importance estimates using PSMA PET and CT radiomic features, and demonstrate a methodology for predicting patient-specific importance deviations from the population. Intra-parotid gland uptake was compared with four regional importance models using 30 [18F]DCFPyL PSMA PET images. A radiomics-based predictive model of population importance was developed using a double cross-validation methodology. Population importance estimates were supplemented using patient-specific radiomic features. Anticorrelative relationships were found to exist between PSMA PET uptake and four independent models of subregional parotid gland importance from the literature. Kernel Ridge Regression with principal component analysis feature selection performed best over test sets (MAE = 0.08), with GLCM features being particularly important. Deblurring PSMA PET images strengthened correlations and improved model performance. This study suggests that regions of relatively low PSMA PET concentration in parotid glands may exhibit relatively high dose-sensitivity. We've demonstrated the utility of PSMA PET radiomic features for predicting relative importance within the parotid glands. PSMA PET appears promising for analyzing salivary gland functionality., Comment: 9 Figures, 7 Tables

There is a need for open-source libraries in emission tomography that (i) use modern and popular backend code to encourage community contributions and (ii) offer support for the multitude of reconstruction techniques available in recent literature, such as those that employ artificial intelligence. The purpose of this research was to create and evaluate a GPU-accelerated, open-source, and user-friendly image reconstruction library, designed to serve as a central platform for the development, validation, and deployment of various tomographic reconstruction algorithms. PyTomography was developed using Python and inherits the GPU-accelerated functionality of PyTorch and parallelproj for fast computations. Its flexible and modular design decouples system matrices, likelihoods, and reconstruction algorithms, simplifying the process of integrating new imaging modalities using various python tools. Example use cases demonstrate the software capabilities in parallel hole SPECT and listmode PET imaging. Overall, we have developed and publicly share PyTomography, a highly optimized and user-friendly software for medical image reconstruction, with a class hierarchy that fosters the development of novel imaging applications., Comment: 28 pages, 5 figures

Objective: To simultaneously deblur and supersample prostate specific membrane antigen (PSMA) positron emission tomography (PET) images using neural blind deconvolution. Approach: Blind deconvolution is a method of estimating the hypothetical "deblurred" image along with the blur kernel (related to the point spread function) simultaneously. Traditional \textit{maximum a posteriori} blind deconvolution methods require stringent assumptions and suffer from convergence to a trivial solution. A method of modelling the deblurred image and kernel with independent neural networks, called "neural blind deconvolution" had demonstrated success for deblurring 2D natural images in 2020. In this work, we adapt neural blind deconvolution for PVE correction of PSMA PET images with simultaneous supersampling. We compare this methodology with several interpolation methods, using blind image quality metrics, and test the model's ability to predict kernels by re-running the model after applying artificial "pseudokernels" to deblurred images. The methodology was tested on a retrospective set of 30 prostate patients as well as phantom images containing spherical lesions of various volumes. Results: Neural blind deconvolution led to improvements in image quality over other interpolation methods in terms of blind image quality metrics, recovery coefficients, and visual assessment. Predicted kernels were similar between patients, and the model accurately predicted several artificially-applied pseudokernels. Localization of activity in phantom spheres was improved after deblurring, allowing small lesions to be more accurately defined. Significance: The intrinsically low spatial resolution of PSMA PET leads to PVEs which negatively impact uptake quantification in small regions. The proposed method can be used to mitigate this issue, and can be straightforwardly adapted for other imaging modalities., Comment: 10 Figures, 4 Tables, 19 pages

This work aims to investigate the accuracy of quantitative SPECT imaging of $^{177}$Lu in the presence of $^{90}$Y, which occurs in dual-isotope radiopharmaceutical therapy (RPT) involving both isotopes. We used the GATE Monte Carlo simulation toolkit to conduct a phantom study, simulating spheres filled with $^{177}$Lu and $^{90}$Y placed in a cylindrical water phantom that was also filled with activity of both radionuclides. We simulated multiple phantom configurations and activity combinations by varying the location of the spheres, the concentrations of $^{177}$Lu and $^{90}$Y in the spheres, and the amount of background activity. We investigated two different scatter window widths to be used with triple energy window (TEW) scatter correction. We also created multiple realizations of each configuration to improve our assessment, leading to a total of 540 simulations. Each configuration was imaged using a simulated Siemens SPECT camera. The projections were reconstructed using the standard 3D OSEM algorithm, and errors associated with $^{177}$Lu activity quantification and contrast-to-noise ratios (CNRs) were determined. In all configurations, the quantification error was within $\pm$6% of the no-$^{90}$Y case, and we found that quantitative accuracy may slightly improve when $^{90}$Y is present because of reduction of errors associated with TEW scatter correction. The CNRs were not significantly impacted by the presence of $^{90}$Y, but they were increased when a wider scatter window width was used for TEW scatter correction. The width of the scatter windows made a small but statistically significant difference of 1-2% on the recovered $^{177}$Lu activity. Based on these results, we can conclude that activity quantification of $^{177}$Lu and lesion detectability is not degraded by the presence of $^{90}$Y., Comment: 16 pages, 7 figures, submitted to Biomedical Physics and Engineering Express

The time-consuming task of manual segmentation challenges routine systematic quantification of disease burden. Convolutional neural networks (CNNs) hold significant promise to reliably identify locations and boundaries of tumors from PET scans. We aimed to leverage the need for annotated data via semi-supervised approaches, with application to PET images of diffuse large B-cell lymphoma (DLBCL) and primary mediastinal large B-cell lymphoma (PMBCL). We analyzed 18F-FDG PET images of 292 patients with PMBCL (n=104) and DLBCL (n=188) (n=232 for training and validation, and n=60 for external testing). We employed FCM and MS losses for training a 3D U-Net with different levels of supervision: i) fully supervised methods with labeled FCM (LFCM) as well as Unified focal and Dice loss functions, ii) unsupervised methods with Robust FCM (RFCM) and Mumford-Shah (MS) loss functions, and iii) Semi-supervised methods based on FCM (RFCM+LFCM), as well as MS loss in combination with supervised Dice loss (MS+Dice). Unified loss function yielded higher Dice score (mean +/- standard deviation (SD)) (0.73 +/- 0.03; 95% CI, 0.67-0.8) compared to Dice loss (p-value<0.01). Semi-supervised (RFCM+alpha*LFCM) with alpha=0.3 showed the best performance, with a Dice score of 0.69 +/- 0.03 (95% CI, 0.45-0.77) outperforming (MS+alpha*Dice) for any supervision level (any alpha) (p<0.01). The best performer among (MS+alpha*Dice) semi-supervised approaches with alpha=0.2 showed a Dice score of 0.60 +/- 0.08 (95% CI, 0.44-0.76) compared to another supervision level in this semi-supervised approach (p<0.01). Semi-supervised learning via FCM loss (RFCM+alpha*LFCM) showed improved performance compared to supervised approaches. Considering the time-consuming nature of expert manual delineations and intra-observer variabilities, semi-supervised approaches have significant potential for automated segmentation workflows.

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

The SNMMI Artificial Intelligence (SNMMI-AI) Summit, organized by the SNMMI AI Task Force, took place in Bethesda, MD on March 21-22, 2022. It brought together various community members and stakeholders from academia, healthcare, industry, patient representatives, and government (NIH, FDA), and considered various key themes to envision and facilitate a bright future for routine, trustworthy use of AI in nuclear medicine. In what follows, essential issues, challenges, controversies and findings emphasized in the meeting are summarized.

Segmentation of lymphoma lesions is challenging due to their varied sizes and locations in whole-body PET scans. This work presents a fully-automated segmentation technique using a multi-center dataset of diffuse large B-cell lymphoma (DLBCL) with heterogeneous characteristics. We utilized a dataset of [18F]FDG-PET scans (n=194) from two different imaging centers, including cases with primary mediastinal large B-cell lymphoma (PMBCL) (n=104). Automated brain and bladder removal approaches were utilized as preprocessing steps to tackle false positives caused by normal hypermetabolic uptake in these organs. Our segmentation model is a convolutional neural network (CNN) based on a 3D U-Net architecture that includes squeeze and excitation (SE) modules. Hybrid distribution, region, and boundary-based losses (Unified Focal and Mumford-Shah (MS)) were utilized that showed the best performance compared to other combinations (p<0.05). Cross-validation between different centers, DLBCL and PMBCL cases, and three random splits were applied on train/validation data. The ensemble of these six models achieved a Dice similarity coefficient (DSC) of 0.77 +- 0.08 and Hausdorff distance (HD) of 16.5 +-12.5. Our 3D U-net model with SE modules for segmentation with hybrid loss performed significantly better (p<0.05) as compared to the 3D U-Net (without SE modules) using the same loss function (Unified Focal and MS loss) (DSC= 0.64 +-0.21 and HD= 26.3 +- 18.7). Our model can facilitate a fully automated quantification pipeline in a multi-center context that opens the possibility for routine reporting of total metabolic tumor volume (TMTV) and other metrics shown useful for the management of lymphoma.

Training computer-vision related algorithms on medical images for disease diagnosis or image segmentation is difficult in large part due to privacy concerns. For this reason, generative image models are highly sought after to facilitate data sharing. However, 3-D generative models are understudied, and investigation of their privacy leakage is needed. We introduce our 3-D generative model, Transversal GAN (TrGAN), using head & neck PET images which are conditioned on tumour masks as a case study. We define quantitative measures of image fidelity, utility and privacy for our model. These metrics are evaluated in the course of training to identify ideal fidelity, utility and privacy trade-offs and establish the relationships between these parameters. We show that the discriminator of the TrGAN is vulnerable to attack, and that an attacker can identify which samples were used in training with almost perfect accuracy (AUC = 0.99). We also show that an attacker with access to only the generator cannot reliably classify whether a sample had been used for training (AUC = 0.51). This suggests that TrGAN generators, but not discriminators, may be used for sharing synthetic 3-D PET data with minimal privacy risk while maintaining good utility and fidelity., Comment: arXiv admin note: text overlap with arXiv:2111.01866

Radiomics features extract quantitative information from medical images, towards the derivation of biomarkers for clinical tasks, such as diagnosis, prognosis, or treatment response assessment. Different image discretization parameters (e.g. bin number or size), convolutional filters, segmentation perturbation, or multi-modality fusion levels can be used to generate radiomics features and ultimately signatures. Commonly, only one set of parameters is used; resulting in only one value or flavour for a given RF. We propose tensor radiomics (TR) where tensors of features calculated with multiple combinations of parameters (i.e. flavours) are utilized to optimize the construction of radiomics signatures. We present examples of TR as applied to PET/CT, MRI, and CT imaging invoking machine learning or deep learning solutions, and reproducibility analyses: (1) TR via varying bin sizes on CT images of lung cancer and PET-CT images of head & neck cancer (HNC) for overall survival prediction. A hybrid deep neural network, referred to as TR-Net, along with two ML-based flavour fusion methods showed improved accuracy compared to regular rediomics features. (2) TR built from different segmentation perturbations and different bin sizes for classification of late-stage lung cancer response to first-line immunotherapy using CT images. TR improved predicted patient responses. (3) TR via multi-flavour generated radiomics features in MR imaging showed improved reproducibility when compared to many single-flavour features. (4) TR via multiple PET/CT fusions in HNC. Flavours were built from different fusions using methods, such as Laplacian pyramids and wavelet transforms. TR improved overall survival prediction. Our results suggest that the proposed TR paradigm has the potential to improve performance capabilities in different medical imaging tasks.

We utilized a 3D nnU-Net model with residual layers supplemented by squeeze and excitation (SE) normalization for tumor segmentation from PET/CT images provided by the Head and Neck Tumor segmentation chal-lenge (HECKTOR). Our proposed loss function incorporates the Unified Fo-cal and Mumford-Shah losses to take the advantage of distribution, region, and boundary-based loss functions. The results of leave-one-out-center-cross-validation performed on different centers showed a segmentation performance of 0.82 average Dice score (DSC) and 3.16 median Hausdorff Distance (HD), and our results on the test set achieved 0.77 DSC and 3.01 HD. Following lesion segmentation, we proposed training a case-control proportional hazard Cox model with an MLP neural net backbone to predict the hazard risk score for each discrete lesion. This hazard risk prediction model (CoxCC) was to be trained on a number of PET/CT radiomic features extracted from the segmented lesions, patient and lesion demographics, and encoder features provided from the penultimate layer of a multi-input 2D PET/CT convolutional neural network tasked with predicting time-to-event for each lesion. A 10-fold cross-validated CoxCC model resulted in a c-index validation score of 0.89, and a c-index score of 0.61 on the HECKTOR challenge test dataset.