Amir Hossein Pouria

Amir Hossein Pouria

Auto-generated profile, based on publicly available information

Amir Hossein Pouria

Department of Computer Engineering, Amirkabir University of Technology, Tehran, Iran

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 January 2026 in Physics in Medicine & Biology
Verified
Journal Article
Open access
AGArman Gorji
Arman Gorji
Nima SanatiNima Sanati
Nima Sanati
APAmir Hossein Pouria
Amir Hossein Pouria
SMSomayeh Sadat Mehrnia
Somayeh Sadat Mehrnia
IHIlker Hacihaliloglu
Ilker Hacihaliloglu
MSMohammad R. Salmanpour
Mohammad R. Salmanpour
Objective: Radiomics-based AI models show potential in breast cancer diagnosis but lack interpretability. This study bridges the gap between radiomic features (RF) and BI-RADS descriptors through a clinically interpretable framework. Methods: We developed a dual-dictionary approach. First, a Clinical Mapping Dictionary (CMD) was constructed by mapping 56 RFs to BI-RADS descriptors (shape, margin, internal enhancement) based on literature and expert review. Second, we applied this framework to a classification task to predict triple-negative (TNBC) versus non-TNBC subtypes using Dynamic Contrast-Enhanced MRI data from a multi-institutional cohort of 1,549 patients. We trained 27 machine learning classifiers with 27 feature selection methods. Using SHapley Additive exPlanations (SHAP), we interpreted the model's predictions and developed a Statistical Mapping Dictionary (SMD) for 51 RFs, not included in the CMD. Results: The best-performing model (Variance Inflation Factor feature selector+Extra Trees Classifier) achieved an average cross-validation accuracy of 0.83±0.02. Our dual-dictionary approach successfully translated predictive RFs into understandable clinical concepts. For example, higher values of 'Sphericity', corresponding to a round/oval shape, were predictive of TNBC. Similarly, lower values of 'Busyness', indicating more homogeneous internal enhancement, were also associated with TNBC, aligning with existing clinical observations. This framework confirmed known imaging biomarkers and identified novel, data-driven quantitative features. Conclusion: This study introduces a novel dual-dictionary framework (BM1.0) that bridges RFs and the BI-RADS clinical lexicon. By enhancing the interpretability and transparency of AI models, the framework supports greater clinical trust and paves the way for integrating RFs into breast cancer diagnosis and personalized care.

24 May 2026 in Computer Methods and Programs in Biomedicine
Verified
Journal Article
Open access
MSMohammad R. Salmanpour
Mohammad R. Salmanpour
APAmir Hossein Pouria
Amir Hossein Pouria
SBSirwan Barichin
Sirwan Barichin
YSYasaman Salehi
Yasaman Salehi
SFSonya Falahati
Sonya Falahati
ARArman Rahmim
Arman Rahmim
Radiomics analyses extract quantitative biomarkers from medical images for precision modeling, yet reproducibility and scalability remain limited by heterogeneous and limited implementations. Existing tools support only partial standards and lack integration with deep learning (DL) radiomics. To address these gaps, we developed PySERA, an open-source, Python-native, standardized radiomics framework designed for automation, reproducibility, and AI integration. PySERA re-implements MATLAB-based SERA (standardized environment for radiomics analysis) in a modular, object-oriented Python architecture. It computes 557 features, including 487 features compliant with the Image Biomarker Standardization Initiative (IBSI) and 10 moment-invariant descriptors, as well as 60 additional diagnostic features, along with DL radiomics embeddings from pre-trained DL: ResNet50 (2048 features) DL radiomics features), DenseNet121 (1024), and VGG16 (512). It includes standardized preprocessing (resampling, discretization, normalization), multi-format I/O (DICOM, NIfTI, NRRD), adaptive memory handling, and a parallel multi-core engine for scalable feature extraction. PySERA integrates directly with libraries: scikit-learn/PyTorch/TensorFlow/MONAI, and others for downstream machine learning applications. PySERA demonstrated >94% IBSI reproducibility, closely matching MITK and substantially outperforming PyRadiomics against the 487 IBSI-compliant feature reference set. Across 8 public datasets, PySERA achieved accuracies of 0.43-0.84, exceeding PyRadiomics for outcome prediction tasks. Benchmarking showed efficient processing (including added higher-order features not implemented in other software): 583 seconds (305 MB) for 166 features, and 2325 seconds (491 MB) for full extraction, with deterministic outputs across platforms. By uniting standardized handcrafted/DL radiomics in a scalable, transparent, and Python-integrable framework, PySERA establishes a reproducible and extensible foundation for next-generation, AI-ready precision imaging research.

10 June 2026 in IEEE Transactions on Biomedical Engineering
Verified
Journal Article
Closed access
MSMohammad R. Salmanpour
Mohammad R. Salmanpour
APAmir Hossein Pouria
Amir Hossein Pouria
SFSonya Falahati
Sonya Falahati
STShahram Taeb
Shahram Taeb
SMSomayeh Sadat Mehrnia
Somayeh Sadat Mehrnia
ARArman Rahmim
Arman Rahmim
Objective: Computed tomography (CT) imaging is essential for lung cancer (LCa) management, while offering detailed visualization and valuable information for Artificial Intelligence-guided prognosis. Meanwhile, supervised learning (SL) models require extensive labeled data, limiting their real-world utility where annotations are scarce. Methods: We analyzed CT scans from 977 patients across 12 public/private datasets, extracting 1,218 radiomics features using Laplacian of Gaussian and wavelet filters via standardized PyRadiomics. Dimensionality was reduced using 56 feature selection and attribute extraction algorithms, and 27 classifiers were benchmarked. Semi-supervised learning (SSL) framework with pseudo-labeling utilized 478 unlabeled and 499 labeled cases. Model sensitivity was assessed in three scenarios: varying labeled data in SL, increasing unlabeled data in SSL, and jointly scaling both from 10% to 100%. SHapley Additive exPlanations analysis (SHAP)analysis explained and interpreted the top model predictions. Five-fold cross-validation and external testing (in two cohorts) were performed. Results: SSL outperformed SL across all metrics, improving overall survival prediction by up to 17%. The top SSL model (Feature Importance by Random Forest+XGBoost) achieved 0.90±0.01 accuracy in cross-validation and 0.88±0.01 externally. SHAP revealed enhanced feature discriminability in SSL and SL, particularly for class 1 (survival>4 years), and helped explain model decisions. SSL maintained strong performance even with only 10% labeled data. Both SSL-based scenarios demonstrated more stable performance compared to SL, with lower variance across external testing, emphasizing SSL's robustness and cost-efficiency. Conclusion: We propose an interpretable, cost-effective SSL framework for CT-based LCa survival prediction. Significance: It improves performance, generalizability, and clinical readiness via SHAP explainability and unlabeled data utilization.