arXiv subjects

Mohammad R. Salmanpour

Publications and source records attributed to Mohammad R. Salmanpour.

18 recordsLinked to original sources

arXiv2025

Towards Interpretable AI in Personalized Medicine: A Radiological-Biological Radiomics Dictionary Connecting Semantic Lung-RADS and imaging Radiomics Features; Dictionary LC 1.0

Lung cancer remains the leading cause of cancer-related mortality worldwide, with survival strongly dependent on early detection. Standard-dose computed tomography (CT) screening using the Lung Imaging Reporting and Data System (Lung-RADS) standardizes pulmonary nodule assessment but is limited by inter-reader variability and reliance on qualitative descriptors, while radiomics offers quantitative biomarkers that often lack clinical interpretability. To bridge this gap, we propose a radiological-biological dictionary that aligns radiomic features (RFs) with Lung-RADS semantic categories. A clinically informed dictionary translating ten Lung-RADS descriptors into radiomic proxies was developed through literature curation and validated by eight expert reviewers. As a proof of concept, imaging and clinical data from 977 patients across 12 collections in The Cancer Imaging Archive (TCIA) were analyzed; following preprocessing and manual segmentation, 110 RFs per nodule were extracted using PyRadiomics in compliance with the Image Biomarker Standardization Initiative (IBSI). A semi-supervised learning framework incorporating 499 labeled and 478 unlabeled cases was applied to improve generalizability, evaluating seven feature selection methods and ten interpretable classifiers. The optimal pipeline (ANOVA feature selection with a support vector machine) achieved a mean validation accuracy of 0.79. SHapley Additive exPlanations (SHAP) analysis identified key RFs corresponding to Lung-RADS semantics such as attenuation, margin irregularity, and spiculation, supporting the validity of the proposed mapping. Overall, this dictionary provides an interpretable framework linking radiomics and Lung-RADS semantics, advancing explainable artificial intelligence for CT-based lung cancer screening.

physics.med-ph↗

arXiv2025

PySERA: Open-Source Standardized Python Library for Automated, Scalable, and Reproducible Handcrafted and Deep Radiomics

Radiomics enables the extraction of quantitative biomarkers from medical images for precision modeling, but reproducibility and scalability remain limited due to heterogeneous software implementations and incomplete adherence to standards. Existing tools also lack unified support for deep learning based radiomics. To address these limitations, we introduce PySERA, an open source, Python native, standardized radiomics framework designed for automation, reproducibility, and seamless AI integration. PySERA reimplements the MATLAB based SERA platform within a modular, object oriented architecture and computes 557 features, including 487 Image Biomarker Standardization Initiative (IBSI) compliant features, 10 moment invariant descriptors, and 60 diagnostic features, together with deep learning radiomics embeddings from pretrained networks such as ResNet50, DenseNet121, and VGG16. The framework provides standardized preprocessing, including resampling, discretization, and normalization, multi format image input and output, adaptive memory management, and a parallel multicores extraction engine. PySERA integrates natively with major machine learning ecosystems including scikit learn, PyTorch, TensorFlow, MONAI and others. In IBSI benchmarks, PySERA achieved more than 94 percent reproducibility and outperformed PyRadiomics while closely matching MITK. Across eight public datasets, it achieved predictive accuracies ranging from 0.54 to 0.87, consistently exceeding PyRadiomics. PySERA unifies handcrafted and deep learning radiomics in a transparent, scalable, and extensible Python framework, establishing a robust foundation for reproducible, AI ready imaging research.

physics.med-ph↗

arXiv2025

Click, Predict, Trust: Clinician-in-the-Loop AI Segmentation for Lung Cancer CT-Based Prognosis within the Knowledge-to-Action Framework

Lung cancer remains the leading cause of cancer mortality, with CT imaging central to screening, prognosis, and treatment. Manual segmentation is variable and time-intensive, while deep learning (DL) offers automation but faces barriers to clinical adoption. Guided by the Knowledge-to-Action framework, this study develops a clinician-in-the-loop DL pipeline to enhance reproducibility, prognostic accuracy, and clinical trust. Multi-center CT data from 999 patients across 12 public datasets were analyzed using five DL models (3D Attention U-Net, ResUNet, VNet, ReconNet, SAM-Med3D), benchmarked against expert contours on whole and click-point cropped images. Segmentation reproducibility was assessed using 497 PySERA-extracted radiomic features via Spearman correlation, ICC, Wilcoxon tests, and MANOVA, while prognostic modeling compared supervised (SL) and semi-supervised learning (SSL) across 38 dimensionality reduction strategies and 24 classifiers. Six physicians qualitatively evaluated masks across seven domains, including clinical meaningfulness, boundary quality, prognostic value, trust, and workflow integration. VNet achieved the best performance (Dice = 0.83, IoU = 0.71), radiomic stability (mean correlation = 0.76, ICC = 0.65), and predictive accuracy under SSL (accuracy = 0.88, F1 = 0.83). SSL consistently outperformed SL across models. Radiologists favored VNet for peritumoral representation and smoother boundaries, preferring AI-generated initial masks for refinement rather than replacement. These results demonstrate that integrating VNet with SSL yields accurate, reproducible, and clinically trusted CT-based lung cancer prognosis, highlighting a feasible path toward physician-centered AI translation.

cs.CV↗

arXiv2025

Semi-Supervised Radiomics for Glioblastoma IDH Mutation: Limited Labels, Data Sensitivity, and SHAP Interpretation

Glioblastoma (GBM) is an aggressive brain tumor in which IDH mutation status is a key prognostic biomarker, but traditional testing requires invasive biopsies, emphasizing the need for non-invasive approaches. In this multi-center study, we analyzed MRI sequences (T1, T2-weighted, contrast-enhanced T1, and FLAIR) from 1,329 patients across eight centers, with IDH labels available for 1,061 cases. A total of 1,223 radiomic features per case were extracted using PyRadiomics with Laplacian of Gaussian and wavelet filters, and both supervised learning (SL) and semi-supervised learning (SSL) frameworks were applied, incorporating 38 feature selection/attribute extraction strategies and 24 classifiers. Five-fold cross-validation was performed on UCSF-PDGM and UPENN datasets, and external validation on IvyGAP, TCGA-LGG, and TCGA-GBM, with SHAP analysis conducted for feature interpretability. Multimodal MRI fusion (T1+T2+T1CE+FLAIR) consistently outperformed single-sequence models. The best SSL model (RFE + SVM) achieved 0.93 cross-validation and 0.75 external accuracy, while the top SL model (RFE + Complement Naive Bayes) reached 0.90 and 0.80, respectively. SSL further demonstrated greater robustness to limited sample sizes, maintaining stable performance compared to SL, and SHAP analysis highlighted the amplified role of first-order Root Mean Square (T1CE) and wavelet-based features, strengthening biomarker interpretability. These findings indicate that SSL enhances accuracy, stability, and interpretability in MRI-based IDH prediction, with multimodal MRI fusion providing the most scalable and reliable strategy for non-invasive biomarker discovery in GBM.

physics.med-ph↗

arXiv2025

Enhancement Without Contrast: Stability-Aware Multicenter Machine Learning for Glioma MRI Imaging

Gadolinium-based contrast agents (GBCAs) are central to glioma imaging but raise safety, cost, and accessibility concerns. Predicting contrast enhancement from non-contrast MRI using machine learning (ML) offers a safer alternative, as enhancement reflects tumor aggressiveness and informs treatment planning. Yet scanner and cohort variability hinder robust model selection. We propose a stability-aware framework to identify reproducible ML pipelines for multicenter prediction of glioma MRI contrast enhancement. We analyzed 1,446 glioma cases from four TCIA datasets (UCSF-PDGM, UPENN-GB, BRATS-Africa, BRATS-TCGA-LGG). Non-contrast T1WI served as input, with enhancement derived from paired post-contrast T1WI. Using PyRadiomics under IBSI standards, 108 features were extracted and combined with 48 dimensionality reduction methods and 25 classifiers, yielding 1,200 pipelines. Rotational validation was trained on three datasets and tested on the fourth. Cross-validation prediction accuracies ranged from 0.91 to 0.96, with external testing achieving 0.87 (UCSF-PDGM), 0.98 (UPENN-GB), and 0.95 (BRATS-Africa), with an average of 0.93. F1, precision, and recall were stable (0.87 to 0.96), while ROC-AUC varied more widely (0.50 to 0.82), reflecting cohort heterogeneity. The MI linked with ETr pipeline consistently ranked highest, balancing accuracy and stability. This framework demonstrates that stability-aware model selection enables reliable prediction of contrast enhancement from non-contrast glioma MRI, reducing reliance on GBCAs and improving generalizability across centers. It provides a scalable template for reproducible ML in neuro-oncology and beyond.

cs.CV↗

arXiv2025

Handcrafted vs. Deep Radiomics vs. Fusion vs. Deep Learning: A Comprehensive Review of Machine Learning -Based Cancer Outcome Prediction in PET and SPECT Imaging

Machine learning (ML), including deep learning (DL) and radiomics-based methods, is increasingly used for cancer outcome prediction with PET and SPECT imaging. However, the comparative performance of handcrafted radiomics features (HRF), deep radiomics features (DRF), DL models, and hybrid fusion approaches remains inconsistent across clinical applications. This systematic review analyzed 226 studies published from 2020 to 2025 that applied ML to PET or SPECT imaging for outcome prediction. Each study was evaluated using a 59-item framework covering dataset construction, feature extraction, validation methods, interpretability, and risk of bias. We extracted key details including model type, cancer site, imaging modality, and performance metrics such as accuracy and area under the curve (AUC). PET-based studies (95%) generally outperformed those using SPECT, likely due to higher spatial resolution and sensitivity. DRF models achieved the highest mean accuracy (0.862), while fusion models yielded the highest AUC (0.861). ANOVA confirmed significant differences in performance (accuracy: p=0.0006, AUC: p=0.0027). Common limitations included inadequate handling of class imbalance (59%), missing data (29%), and low population diversity (19%). Only 48% of studies adhered to IBSI standards. These findings highlight the need for standardized pipelines, improved data quality, and explainable AI to support clinical integration.

physics.med-ph↗

arXiv2025

Radiological and Biological Dictionary of Radiomics Features: Addressing Understandable AI Issues in Personalized Breast Cancer; Dictionary Version BM1.0

Radiomics-based AI models show promise for breast cancer diagnosis but often lack interpretability, limiting clinical adoption. This study addresses the gap between radiomic features (RF) and the standardized BI-RADS lexicon by proposing a dual-dictionary framework. First, a Clinically-Informed Feature Interpretation Dictionary (CIFID) was created by mapping 56 RFs to BI-RADS descriptors (shape, margin, internal enhancement) through literature and expert review. The framework was applied to classify triple-negative breast cancer (TNBC) versus non-TNBC using dynamic contrast-enhanced MRI from a multi-institutional cohort of 1,549 patients. We trained 27 machine learning classifiers with 27 feature selection methods. SHapley Additive exPlanations (SHAP) were used to interpret predictions and generate a complementary Data-Driven Feature Interpretation Dictionary (DDFID) for 52 additional RFs. The best model, combining Variance Inflation Factor (VIF) selection with Extra Trees Classifier, achieved an average cross-validation accuracy of 0.83. Key predictive RFs aligned with clinical knowledge: higher Sphericity (round/oval shape) and lower Busyness (more homogeneous enhancement) were associated with TNBC. The framework confirmed known imaging biomarkers and uncovered novel, interpretable associations. This dual-dictionary approach (BM1.0) enhances AI model transparency and supports the integration of RFs into routine breast cancer diagnosis and personalized care.

physics.comp-ph↗

arXiv2025

Influence of High-Performance Image-to-Image Translation Networks on Clinical Visual Assessment and Outcome Prediction: Utilizing Ultrasound to MRI Translation in Prostate Cancer

Purpose: This study examines the core traits of image-to-image translation (I2I) networks, focusing on their effectiveness and adaptability in everyday clinical settings. Methods: We have analyzed data from 794 patients diagnosed with prostate cancer (PCa), using ten prominent 2D/3D I2I networks to convert ultrasound (US) images into MRI scans. We also introduced a new analysis of Radiomic features (RF) via the Spearman correlation coefficient to explore whether networks with high performance (SSIM>85%) could detect subtle RFs. Our study further examined synthetic images by 7 invited physicians. As a final evaluation study, we have investigated the improvement that are achieved using the synthetic MRI data on two traditional machine learning and one deep learning method. Results: In quantitative assessment, 2D-Pix2Pix network substantially outperformed the other 7 networks, with an average SSIM~0.855. The RF analysis revealed that 76 out of 186 RFs were identified using the 2D-Pix2Pix algorithm alone, although half of the RFs were lost during the translation process. A detailed qualitative review by 7 medical doctors noted a deficiency in low-level feature recognition in I2I tasks. Furthermore, the study found that synthesized image-based classification outperformed US image-based classification with an average accuracy and AUC~0.93. Conclusion: This study showed that while 2D-Pix2Pix outperformed cutting-edge networks in low-level feature discovery and overall error and similarity metrics, it still requires improvement in low-level feature performance, as highlighted by Group 3. Further, the study found using synthetic image-based classification outperformed original US image-based methods.

eess.IV↗

arXiv2025

Robust Semi-Supervised CT Radiomics for Lung Cancer Prognosis: Cost-Effective Learning with Limited Labels and SHAP Interpretation

Background: CT imaging is vital for lung cancer management, offering detailed visualization for AI-based prognosis. However, supervised learning SL models require large labeled datasets, limiting their real-world application in settings with scarce annotations. Methods: We analyzed CT scans from 977 patients across 12 datasets extracting 1218 radiomics features using Laplacian of Gaussian and wavelet filters via PyRadiomics Dimensionality reduction was applied with 56 feature selection and extraction algorithms and 27 classifiers were benchmarked A semi supervised learning SSL framework with pseudo labeling utilized 478 unlabeled and 499 labeled cases Model sensitivity was tested in three scenarios varying labeled data in SL increasing unlabeled data in SSL and scaling both from 10 percent to 100 percent SHAP analysis was used to interpret predictions Cross validation and external testing in two cohorts were performed. Results: SSL outperformed SL, improving overall survival prediction by up to 17 percent. The top SSL model, Random Forest plus XGBoost classifier, achieved 0.90 accuracy in cross-validation and 0.88 externally. SHAP analysis revealed enhanced feature discriminability in both SSL and SL, especially for Class 1 survival greater than 4 years. SSL showed strong performance with only 10 percent labeled data, with more stable results compared to SL and lower variance across external testing, highlighting SSL's robustness and cost effectiveness. Conclusion: We introduced a cost-effective, stable, and interpretable SSL framework for CT-based survival prediction in lung cancer, improving performance, generalizability, and clinical readiness by integrating SHAP explainability and leveraging unlabeled data.

physics.med-ph↗

arXiv2025

Censor-Aware Semi-Supervised Survival Time Prediction in Lung Cancer Using Clinical and Radiomics Features

Objectives: Lung cancer poses a significant global health challenge, necessitating improved prognostic methods for personalized treatment. This study introduces a censor-aware semi-supervised learning (SSL) framework that integrates clinical and imaging data, addressing biases in traditional models handling censored data. Methods: We analyzed clinical, PET and CT data from 199 lung cancer patients from public and local data respositories, focusing on overall survival (OS) time as the primary outcome. Handcrafted (HRF) and Deep Radiomics features (DRF) were extracted after preprocessing using ViSERA software and were combined with clinical features (CF). Feature dimensions were optimized using Principal Component Analysis (PCA), followed by the application of supervised learning (SL) and SSL. SSL incorporated pseudo-labeling of censored data to improve performance. Seven regressors and three hazard ratio survival analysis (HRSA) algorithms were optimized using five-fold cross-validation, grid search and external test bootstrapping. Results: For PET HRFs, SSL reduced the mean absolute error (MAE) by 26.5%, achieving 1.55 years with PCA+decision tree regression, compared to SL's 2.11 years with PCA+KNNR (p<0.05). Combining HRFs (CT_HRF) and DRFs from CT images using SSL+PCA+KNNR achieved an MAE of 2.08 years, outperforming SL's 2.26 years by 7.96% (p<0.05). In HRSA, CT_HRF applied to PCA+Component Wise Gradient Boosting Survival Analysis achieved an external c-index of 0.65, effectively differentiating high- and low-risk groups. Conclusions: We demonstrated that the SSL strategy significantly outperforms SL across PET, CT, and CF. As such, censor-aware SSL applied to HRFs from PET images significantly improved survival prediction performance by 26.5% compared to the SL approach.

physics.med-ph↗

arXiv2025

An AI-powered Public Health Automated Kiosk System for Personalized Care: An Experimental Pilot Study

Background: The HERMES Kiosk (Healthcare Enhanced Recommendations through Artificial Intelligence & Expertise System) is designed to provide personalized Over-the-Counter (OTC) medication recommendations, addressing the limitations of traditional health kiosks. It integrates an advanced GAMENet model enhanced with Graph Attention Networks (GAT) and Multi-Head Cross-Attention (MHCA) while ensuring user privacy through federated learning. This paper outlines the conceptual design and architecture of HERMES, with a focus on deployment in high-traffic public areas. Methods: HERMES analyzes self-reported symptoms and anonymized medical histories using AI algorithms to generate context-aware OTC medication recommendations. The system was initially trained using Electronic Health Records (EHR) from the MIMIC-III dataset (6,350 patients) and Drug-Drug Interaction (DDI) data from the TWOSIDES database, incorporating the top 90 severity DDI types. Real-time DDI checks and ATC-mapped drug codes further improve safety. The kiosk is designed for accessibility, offering multilingual support, large fonts, voice commands, and Braille compatibility. A built-in health education library promotes preventive care and health literacy. A survey was conducted among 10 medical professionals to evaluate its potential applications in medicine. Results: Preliminary results show that the enhanced GAMENet model achieved a Precision-Recall AUC (PRAUC) of 0.74, outperforming the original model. These findings suggest a strong potential for delivering accurate and secure healthcare recommendations in public settings. Conclusion: HERMES demonstrates how AI-driven, privacy-preserving kiosks can enhance public health access, empower users, and alleviate burdens on healthcare systems. Future work will focus on real-world deployment, usability testing, and scalability for broader adoption.

cs.HC↗

arXiv2025

AllMetrics: A Unified Python Library for Standardized Metric Evaluation and Robust Data Validation in Machine Learning

Machine learning (ML) models rely heavily on consistent and accurate performance metrics to evaluate and compare their effectiveness. However, existing libraries often suffer from fragmentation, inconsistent implementations, and insufficient data validation protocols, leading to unreliable results. Existing libraries have often been developed independently and without adherence to a unified standard, particularly concerning the specific tasks they aim to support. As a result, each library tends to adopt its conventions for metric computation, input/output formatting, error handling, and data validation protocols. This lack of standardization leads to both implementation differences (ID) and reporting differences (RD), making it difficult to compare results across frameworks or ensure reliable evaluations. To address these issues, we introduce AllMetrics, an open-source unified Python library designed to standardize metric evaluation across diverse ML tasks, including regression, classification, clustering, segmentation, and image-to-image translation. The library implements class-specific reporting for multi-class tasks through configurable parameters to cover all use cases, while incorporating task-specific parameters to resolve metric computation discrepancies across implementations. Various datasets from domains like healthcare, finance, and real estate were applied to our library and compared with Python, Matlab, and R components to identify which yield similar results. AllMetrics combines a modular Application Programming Interface (API) with robust input validation mechanisms to ensure reproducibility and reliability in model evaluation. This paper presents the design principles, architectural components, and empirical analyses demonstrating the ability to mitigate evaluation errors and to enhance the trustworthiness of ML workflows.

cs.LG↗

arXiv2025

Pathobiological Dictionary Defining Pathomics and Texture Features: Addressing Understandable AI Issues in Personalized Liver Cancer; Dictionary Version LCP1.0

Artificial intelligence (AI) holds strong potential for medical diagnostics, yet its clinical adoption is limited by a lack of interpretability and generalizability. This study introduces the Pathobiological Dictionary for Liver Cancer (LCP1.0), a practical framework designed to translate complex Pathomics and Radiomics Features (PF and RF) into clinically meaningful insights aligned with existing diagnostic workflows. QuPath and PyRadiomics, standardized according to IBSI guidelines, were used to extract 333 imaging features from hepatocellular carcinoma (HCC) tissue samples, including 240 PF-based-cell detection/intensity, 74 RF-based texture, and 19 RF-based first-order features. Expert-defined ROIs from the public dataset excluded artifact-prone areas, and features were aggregated at the case level. Their relevance to the WHO grading system was assessed using multiple classifiers linked with feature selectors. The resulting dictionary was validated by 8 experts in oncology and pathology. In collaboration with 10 domain experts, we developed a Pathobiological dictionary of imaging features such as PFs and RF. In our study, the Variable Threshold feature selection algorithm combined with the SVM model achieved the highest accuracy (0.80, P-value less than 0.05), selecting 20 key features, primarily clinical and pathomics traits such as Centroid, Cell Nucleus, and Cytoplasmic characteristics. These features, particularly nuclear and cytoplasmic, were strongly associated with tumor grading and prognosis, reflecting atypia indicators like pleomorphism, hyperchromasia, and cellular orientation.The LCP1.0 provides a clinically validated bridge between AI outputs and expert interpretation, enhancing model transparency and usability. Aligning AI-derived features with clinical semantics supports the development of interpretable, trustworthy diagnostic tools for liver cancer pathology.

physics.comp-ph↗

arXiv2025

Machine Learning Evaluation Metric Discrepancies across Programming Languages and Their Components: Need for Standardization

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.

cs.LG↗

arXiv2024

Biological and Radiological Dictionary of Radiomics Features: Addressing Understandable AI Issues in Personalized Prostate Cancer; Dictionary Version PM1.0

We investigate the connection between visual semantic features defined in PI-RADS and associated risk factors, moving beyond abnormal imaging findings, establishing a shared framework between medical and AI professionals by creating a standardized dictionary of biological/radiological RFs. Subsequently, 6 interpretable and seven complex classifiers, linked with nine interpretable feature selection algorithms (FSA) applied to risk factors, were extracted from segmented lesions in T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) multiparametric-prostate MRI sequences to predict the UCLA scores. We then utilized the created dictionary to interpret the best-predictive models. Combining T2WI, DWI, and ADC with FSAs including ANOVA F-test, Correlation Coefficient, and Fisher Score, and utilizing logistic regression, identified key features: The 90th percentile from T2WI, which captures hypo-intensity related to prostate cancer risk; Variance from T2WI, indicating lesion heterogeneity; shape metrics including Least Axis Length and Surface Area to Volume ratio from ADC, describing lesion shape and compactness; and Run Entropy from ADC, reflecting texture consistency. This approach achieved the highest average accuracy of 0.78, significantly outperforming single-sequence methods (p-value<0.05). The developed dictionary for Prostate-MRI (PM1.0) serves as a common language, fosters collaboration between clinical professionals and AI developers to advance trustworthy AI solutions that support reliable/interpretable clinical decisions.

physics.med-ph↗

arXiv2024

Enhanced Lung Cancer Survival Prediction using Semi-Supervised Pseudo-Labeling and Learning from Diverse PET/CT Datasets

Objective: This study explores a semi-supervised learning (SSL), pseudo-labeled strategy using diverse datasets to enhance lung cancer (LCa) survival predictions, analyzing Handcrafted and Deep Radiomic Features (HRF/DRF) from PET/CT scans with Hybrid Machine Learning Systems (HMLS). Methods: We collected 199 LCa patients with both PET & CT images, obtained from The Cancer Imaging Archive (TCIA) and our local database, alongside 408 head&neck cancer (HNCa) PET/CT images from TCIA. We extracted 215 HRFs and 1024 DRFs by PySERA and a 3D-Autoencoder, respectively, within the ViSERA software, from segmented primary tumors. The supervised strategy (SL) employed a HMLSs: PCA connected with 4 classifiers on both HRF and DRFs. SSL strategy expanded the datasets by adding 408 pseudo-labeled HNCa cases (labeled by Random Forest algorithm) to 199 LCa cases, using the same HMLSs techniques. Furthermore, Principal Component Analysis (PCA) linked with 4 survival prediction algorithms were utilized in survival hazard ratio analysis. Results: SSL strategy outperformed SL method (p-value<0.05), achieving an average accuracy of 0.85 with DRFs from PET and PCA+ Multi-Layer Perceptron (MLP), compared to 0.65 for SL strategy using DRFs from CT and PCA+ K-Nearest Neighbor (KNN). Additionally, PCA linked with Component-wise Gradient Boosting Survival Analysis on both HRFs and DRFs, as extracted from CT, had an average c-index of 0.80 with a Log Rank p-value<<0.001, confirmed by external testing. Conclusions: Shifting from HRFs and SL to DRFs and SSL strategies, particularly in contexts with limited data points, enabling CT or PET alone to significantly achieve high predictive performance.

cs.CV↗

arXiv2024

Do High-Performance Image-to-Image Translation Networks Enable the Discovery of Radiomic Features? Application to MRI Synthesis from Ultrasound in Prostate Cancer

This study investigates the foundational characteristics of image-to-image translation networks, specifically examining their suitability and transferability within the context of routine clinical environments, despite achieving high levels of performance, as indicated by a Structural Similarity Index (SSIM) exceeding 0.95. The evaluation study was conducted using data from 794 patients diagnosed with Prostate cancer. To synthesize MRI from Ultrasound images, we employed five widely recognized image to image translation networks in medical imaging: 2DPix2Pix, 2DCycleGAN, 3DCycleGAN, 3DUNET, and 3DAutoEncoder. For quantitative assessment, we report four prevalent evaluation metrics Mean Absolute Error, Mean Square Error, Structural Similarity Index (SSIM), and Peak Signal to Noise Ratio. Moreover, a complementary analysis employing Radiomic features (RF) via Spearman correlation coefficient was conducted to investigate, for the first time, whether networks achieving high performance, SSIM greater than 0.85, could identify low-level RFs. The RF analysis showed 75 features out of 186 RFs were discovered via just 2DPix2Pix algorithm while half of RFs were lost in the translation process. Finally, a detailed qualitative assessment by five medical doctors indicated a lack of low level feature discovery in image to image translation tasks.

eess.IV↗

arXiv2022

Tensor Radiomics: Paradigm for Systematic Incorporation of Multi-Flavoured Radiomics Features

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.

cs.CV↗