Khaled Sayed
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Khaled Sayed has a total of 85 co-authors. Here are the top co-authors by number of shared publications:
- Panayiotis V. Benos
Department of Epidemiology, University of Florida
5 shared publications Aaron Bell
Division of Experimental Pathology, Department of Pathology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA; Pittsburgh Liver Research Center, University of Pittsburgh and UPMC, Pittsburgh, PA, USA
2 shared publications- Abdus S. Wahed
Department of Biostatistics and Computational Biology, University of Rochester, Rochester, New York, USA
2 shared publications Alan Watson
Center for Biologic Imaging, University of Pittsburgh, Pittsburgh, PA, USA; Department of Cell Biology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
2 shared publicationsAndrew Feranchak
Division of Gastroenterology, Hepatology, and Nutrition, Department of Pediatrics, University of Pittsburgh School of Medicine and UPMC, Pittsburgh, PA, USA; Pittsburgh Liver Research Center, University of Pittsburgh and UPMC, Pittsburgh, PA, USA
2 shared publications- Daniel W. Wilkey
University of Louisville
2 shared publications
Latest Publications
- Open Access
- 10 July 2026
- Published by Open Engineering Inc in engrXiv
Grounding Large Language Models (LLMs) in Agency Standards: A Controlled Comparison of Closed-Book, Retrieval-Augmented, and Full-Document Question Answering for Construction Inspection
Objectives: Construction inspectors on transportation projects must verify work against a fragmented body of specifications under schedule pressure. Transportation agencies are now deciding how, and whether, to deploy Large Language Models (LLMs) to support this task. This study examines whether an LLM must be grounded in agency documents at all, and if so, whether grounding should take the form of Retrieval-Augmented Generation (RAG) or simply providing the full source document within the model's context window. Methods: The effect of document grounding was isolated through a controlled, within-question comparison. Three contemporary LLMs from different model families each answered 70 validated inspection questions, drawn from seven public Connecticut Department of Transportation (CTDOT), Occupational Safety and Health Administration (OSHA), and Federal Highway Administration (FHWA) documents, under three conditions: closed-book, RAG, and full-document context. Responses were scored with token-level F1, ROUGE-L, and semantic similarity, together with an LLM-judge rubric validated against blinded ratings by a licensed professional engineer. Consistency across systems was quantified with the Gini coefficient. Findings: Grounding proved decisive: relative to closed-book prompting, RAG raised the judge composite by 1.31 to 1.48 points on the five-point scale and full-document context by 1.40 to 1.66 points ( p < 0.000001), while 56 to 64 percent of closed-book answers were judged factually unacceptable. The two grounding strategies were comparable, with full-document context ahead by average 0.14 points despite requiring roughly 35 times the input tokens. Grounding gains were largest for project-specific documents and smallest for widely published standards (r = -0.83). Novelty: This is the first controlled study to isolate the causal effect of document grounding on transportation inspection question answering, and to compare RAG against full-document prompting for agency standards. Practical Applications: The findings offer agencies evidence-based guidance on when retrieval infrastructure is warranted for inspection support and when simpler deployment suffices.
- 10 July 2026
- Published by Open Engineering Inc in engrXiv
Open AccessGrounding Large Language Models (LLMs) in Agency Standards: A Controlled Comparison of Closed-Book, Retrieval-Augmented, and Full-Document Question Answering for Construction Inspection
Objectives: Construction inspectors on transportation projects must verify work against a fragmented body of specifications under schedule pressure. Transportation agencies are now deciding how, and whether, to deploy Large Language Models (LLMs) to support this task. This study examines whether an LLM must be grounded in agency documents at all, and if so, whether grounding should take the form of Retrieval-Augmented Generation (RAG) or simply providing the full source document within the model's context window. Methods: The effect of document grounding was isolated through a controlled, within-question comparison. Three contemporary LLMs from different model families each answered 70 validated inspection questions, drawn from seven public Connecticut Department of Transportation (CTDOT), Occupational Safety and Health Administration (OSHA), and Federal Highway Administration (FHWA) documents, under three conditions: closed-book, RAG, and full-document context. Responses were scored with token-level F1, ROUGE-L, and semantic similarity, together with an LLM-judge rubric validated against blinded ratings by a licensed professional engineer. Consistency across systems was quantified with the Gini coefficient. Findings: Grounding proved decisive: relative to closed-book prompting, RAG raised the judge composite by 1.31 to 1.48 points on the five-point scale and full-document context by 1.40 to 1.66 points ( p < 0.000001), while 56 to 64 percent of closed-book answers were judged factually unacceptable. The two grounding strategies were comparable, with full-document context ahead by average 0.14 points despite requiring roughly 35 times the input tokens. Grounding gains were largest for project-specific documents and smallest for widely published standards (r = -0.83). Novelty: This is the first controlled study to isolate the causal effect of document grounding on transportation inspection question answering, and to compare RAG against full-document prompting for agency standards. Practical Applications: The findings offer agencies evidence-based guidance on when retrieval infrastructure is warranted for inspection support and when simpler deployment suffices.
- 20 January 2026
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 39 (5) , 4343-4354
Customized CNN Architectures Outperform Pre-Trained Models in Differentiating Normal Brain Tissues, Glioma, Meningioma, and Pituitary Tumors
Early and accurate detection of brain tumors is essential for improving treatment outcomes and patient survival. While pre-trained deep learning models such as ResNet, VGG, and MobileNet have achieved notable success in medical image classification, their generalized architectures often struggle to capture the intricate heterogeneity of brain tissues. This study introduces a customized Convolutional Neural Network (CNN) specifically designed for brain tumor classification, demonstrating superior performance over widely used pre-trained models. The primary objective of this research is to evaluate the diagnostic performance of the customized CNN and pre-trained models in real-world scenarios and establish benchmarks for their accuracy, reliability, and computational efficiency. Furthermore, this study aims to explore and implement optimization techniques that enhance the diagnostic capabilities of the models under investigation. The proposed CNN incorporates optimized convolutional blocks, adaptive fine-tuning, and an advanced data augmentation pipeline to enhance feature extraction and minimize overfitting. When evaluated on the CE-MRI Figshare dataset containing 3064 T1-weighted contrast-enhanced images, the model achieved a validation accuracy of 97.01%, outperforming ResNet50 (89.15%), MobileNetV2 (92.89%), and VGG16 (96.76%). Furthermore, the CNN exhibited strong consistency across all tumor categories—glioma, meningioma, and pituitary tumors—proving its robustness in real-world diagnostic scenarios. These findings confirm that a well-optimized CNN architecture can outperform generic pre-trained models, underscoring the importance of task-specific deep learning designs for medical imaging applications.
- 20 January 2026
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 39 (5) , 4343-4354
Customized CNN Architectures Outperform Pre-Trained Models in Differentiating Normal Brain Tissues, Glioma, Meningioma, and Pituitary Tumors
Early and accurate detection of brain tumors is essential for improving treatment outcomes and patient survival. While pre-trained deep learning models such as ResNet, VGG, and MobileNet have achieved notable success in medical image classification, their generalized architectures often struggle to capture the intricate heterogeneity of brain tissues. This study introduces a customized Convolutional Neural Network (CNN) specifically designed for brain tumor classification, demonstrating superior performance over widely used pre-trained models. The primary objective of this research is to evaluate the diagnostic performance of the customized CNN and pre-trained models in real-world scenarios and establish benchmarks for their accuracy, reliability, and computational efficiency. Furthermore, this study aims to explore and implement optimization techniques that enhance the diagnostic capabilities of the models under investigation. The proposed CNN incorporates optimized convolutional blocks, adaptive fine-tuning, and an advanced data augmentation pipeline to enhance feature extraction and minimize overfitting. When evaluated on the CE-MRI Figshare dataset containing 3064 T1-weighted contrast-enhanced images, the model achieved a validation accuracy of 97.01%, outperforming ResNet50 (89.15%), MobileNetV2 (92.89%), and VGG16 (96.76%). Furthermore, the CNN exhibited strong consistency across all tumor categories—glioma, meningioma, and pituitary tumors—proving its robustness in real-world diagnostic scenarios. These findings confirm that a well-optimized CNN architecture can outperform generic pre-trained models, underscoring the importance of task-specific deep learning designs for medical imaging applications.
- 17 September 2025
- Published by Elsevier in Journal of Microbiological Methods
- Vol. 238, 107267
Integrating multi-compartment microbiome data with clinical parameters enhances mortality prediction using autoencoder
- 17 September 2025
- Published by Elsevier in Journal of Microbiological Methods
- Vol. 238, 107267
Integrating multi-compartment microbiome data with clinical parameters enhances mortality prediction using autoencoder
- 3 July 2025
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 39 (2) , 1274-1288
Integrating MobileNetV3 and SqueezeNet for Multi-class Brain Tumor Classification
Brain tumors pose a critical health threat requiring timely and accurate classification for effective treatment. Traditional MRI analysis is labor-intensive and prone to variability, necessitating reliable automated solutions. This study explores lightweight deep learning models for multi-class brain tumor classification across four categories: glioma, meningioma, pituitary tumors, and no tumor. We investigate the performance of MobileNetV3 and SqueezeNet individually, and a feature-fusion hybrid model that combines their embedding layers. We utilized a publicly available MRI dataset containing 7023 images with a consistent internal split (65% training, 17% validation, 18% test) to ensure reliable evaluation. MobileNetV3 offers deep semantic understanding through its expressive features, while SqueezeNet provides minimal computational overhead. Their feature-level integration creates a balanced approach between diagnostic accuracy and deployment efficiency. Experiments conducted with consistent hyperparameters and preprocessing showed MobileNetV3 achieved the highest test accuracy (99.31%) while maintaining a low parameter count (3.47M), making it suitable for real-world deployment. Grad-CAM visualizations were employed for model explainability, highlighting tumor-relevant regions and helping visualize the specific areas contributing to predictions. Our proposed models outperform several baseline architectures like VGG16 and InceptionV3, achieving high accuracy with significantly fewer parameters. These results demonstrate that well-optimized lightweight networks can deliver accurate and interpretable brain tumor classification.
- 3 July 2025
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 39 (2) , 1274-1288
Integrating MobileNetV3 and SqueezeNet for Multi-class Brain Tumor Classification
Brain tumors pose a critical health threat requiring timely and accurate classification for effective treatment. Traditional MRI analysis is labor-intensive and prone to variability, necessitating reliable automated solutions. This study explores lightweight deep learning models for multi-class brain tumor classification across four categories: glioma, meningioma, pituitary tumors, and no tumor. We investigate the performance of MobileNetV3 and SqueezeNet individually, and a feature-fusion hybrid model that combines their embedding layers. We utilized a publicly available MRI dataset containing 7023 images with a consistent internal split (65% training, 17% validation, 18% test) to ensure reliable evaluation. MobileNetV3 offers deep semantic understanding through its expressive features, while SqueezeNet provides minimal computational overhead. Their feature-level integration creates a balanced approach between diagnostic accuracy and deployment efficiency. Experiments conducted with consistent hyperparameters and preprocessing showed MobileNetV3 achieved the highest test accuracy (99.31%) while maintaining a low parameter count (3.47M), making it suitable for real-world deployment. Grad-CAM visualizations were employed for model explainability, highlighting tumor-relevant regions and helping visualize the specific areas contributing to predictions. Our proposed models outperform several baseline architectures like VGG16 and InceptionV3, achieving high accuracy with significantly fewer parameters. These results demonstrate that well-optimized lightweight networks can deliver accurate and interpretable brain tumor classification.
- Open Access
- 30 May 2025
- Published by Elsevier in CHEST Pulmonary
- Vol. 3 (4)
Rapid Metagenomic Sequencing of Bronchoalveolar Lavage Fluid for Diagnosis of Infection in Patients With Hematologic Malignancies and Pulmonary Complications
- 30 May 2025
- Published by Elsevier in CHEST Pulmonary
- Vol. 3 (4)
Open AccessRapid Metagenomic Sequencing of Bronchoalveolar Lavage Fluid for Diagnosis of Infection in Patients With Hematologic Malignancies and Pulmonary Complications
- 23 August 2024
- Published by Springer Nature
- p. 142-151
Medical Equipment Real-Time Locating System in Hospitals Based on Bluetooth Low Energy
- 23 August 2024
- Published by Springer Nature
- p. 142-151
Medical Equipment Real-Time Locating System in Hospitals Based on Bluetooth Low Energy
- Open Access
- 31 July 2024
- Published by Wolters Kluwer Health in Hepatology Communications
- Vol. 8 (8)
A plasma peptidomic signature reveals extracellular matrix remodeling and predicts prognosis in alcohol-associated hepatitis
Background: Alcohol-associated hepatitis (AH) is plagued with high mortality and difficulty in identifying at-risk patients. The extracellular matrix undergoes significant remodeling during inflammatory liver injury and could potentially be used for mortality prediction. Methods: EDTA plasma samples were collected from patients with AH (n = 62); Model for End-Stage Liver Disease score defined AH severity as moderate (12–20; n = 28) and severe (>20; n = 34). The peptidome data were collected by high resolution, high mass accuracy UPLC-MS. Univariate and multivariate analyses identified differentially abundant peptides, which were used for Gene Ontology, parent protein matrisomal composition, and protease involvement. Machine-learning methods were used to develop mortality predictors. Results: Analysis of plasma peptides from patients with AH and healthy controls identified over 1600 significant peptide features corresponding to 130 proteins. These were enriched for extracellular matrix fragments in AH samples, likely related to the turnover of hepatic-derived proteins. Analysis of moderate versus severe AH peptidomes was dominated by changes in peptides from collagen 1A1 and fibrinogen A proteins. The dominant proteases for the AH peptidome spectrum appear to be CAPN1 and MMP12. Causal graphical modeling identified 3 peptides directly linked to 90-day mortality in >90% of the learned graphs. These peptides improved the accuracy of mortality prediction over the Model for End-Stage Liver Disease score and were used to create a clinically applicable mortality prediction assay. Conclusions: A signature based on plasma peptidome is a novel, noninvasive method for prognosis stratification in patients with AH. Our results could also lead to new mechanistic and/or surrogate biomarkers to identify new AH mechanisms. Export
- 31 July 2024
- Published by Wolters Kluwer Health in Hepatology Communications
- Vol. 8 (8)
Open AccessA plasma peptidomic signature reveals extracellular matrix remodeling and predicts prognosis in alcohol-associated hepatitis
Background: Alcohol-associated hepatitis (AH) is plagued with high mortality and difficulty in identifying at-risk patients. The extracellular matrix undergoes significant remodeling during inflammatory liver injury and could potentially be used for mortality prediction. Methods: EDTA plasma samples were collected from patients with AH (n = 62); Model for End-Stage Liver Disease score defined AH severity as moderate (12–20; n = 28) and severe (>20; n = 34). The peptidome data were collected by high resolution, high mass accuracy UPLC-MS. Univariate and multivariate analyses identified differentially abundant peptides, which were used for Gene Ontology, parent protein matrisomal composition, and protease involvement. Machine-learning methods were used to develop mortality predictors. Results: Analysis of plasma peptides from patients with AH and healthy controls identified over 1600 significant peptide features corresponding to 130 proteins. These were enriched for extracellular matrix fragments in AH samples, likely related to the turnover of hepatic-derived proteins. Analysis of moderate versus severe AH peptidomes was dominated by changes in peptides from collagen 1A1 and fibrinogen A proteins. The dominant proteases for the AH peptidome spectrum appear to be CAPN1 and MMP12. Causal graphical modeling identified 3 peptides directly linked to 90-day mortality in >90% of the learned graphs. These peptides improved the accuracy of mortality prediction over the Model for End-Stage Liver Disease score and were used to create a clinically applicable mortality prediction assay. Conclusions: A signature based on plasma peptidome is a novel, noninvasive method for prognosis stratification in patients with AH. Our results could also lead to new mechanistic and/or surrogate biomarkers to identify new AH mechanisms. Export
- Open Access
- 29 July 2024
- Published by MDPI AG in Bioengineering
- Vol. 11 (8) , 764
An Innovative Thermal Imaging Prototype for Precise Breast Cancer Detection: Integrating Compression Techniques and Classification Methods
Breast cancer detection at an early stage is crucial for improving patient survival rates. This work introduces an innovative thermal imaging prototype that incorporates compression techniques inspired by mammography equipment. The prototype offers a radiation-free and precise cancer diagnosis. By integrating compression and illumination methods, thermal picture quality has increased, and the accuracy of classification has improved. Essential components of the suggested thermography device include an equipment body, plates, motors, pressure sensors, light sources, and a thermal camera. We created a 3D model of the gadget using the SolidWorks software 2020 package. Furthermore, the classification research employed both cancer and normal images from the experimental results to validate the efficacy of the suggested system. We employed preprocessing and segmentation methods on the obtained dataset. We successfully categorized the thermal pictures using various classifiers and examined their performance. The logistic regression model showed excellent performance, achieving an accuracy of 0.976, F1 score of 0.977, precision of 1.000, and recall of 0.995. This indicates a high level of accuracy in correctly classifying thermal abnormalities associated with breast cancer. The proposed prototype serves as a highly effective tool for conducting initial investigations into breast cancer detection, offering potential advancements in early-stage diagnosis, and improving patient survival rates.
- 29 July 2024
- Published by MDPI AG in Bioengineering
- Vol. 11 (8) , 764
Open AccessAn Innovative Thermal Imaging Prototype for Precise Breast Cancer Detection: Integrating Compression Techniques and Classification Methods
Breast cancer detection at an early stage is crucial for improving patient survival rates. This work introduces an innovative thermal imaging prototype that incorporates compression techniques inspired by mammography equipment. The prototype offers a radiation-free and precise cancer diagnosis. By integrating compression and illumination methods, thermal picture quality has increased, and the accuracy of classification has improved. Essential components of the suggested thermography device include an equipment body, plates, motors, pressure sensors, light sources, and a thermal camera. We created a 3D model of the gadget using the SolidWorks software 2020 package. Furthermore, the classification research employed both cancer and normal images from the experimental results to validate the efficacy of the suggested system. We employed preprocessing and segmentation methods on the obtained dataset. We successfully categorized the thermal pictures using various classifiers and examined their performance. The logistic regression model showed excellent performance, achieving an accuracy of 0.976, F1 score of 0.977, precision of 1.000, and recall of 0.995. This indicates a high level of accuracy in correctly classifying thermal abnormalities associated with breast cancer. The proposed prototype serves as a highly effective tool for conducting initial investigations into breast cancer detection, offering potential advancements in early-stage diagnosis, and improving patient survival rates.
- Open Access
- 1 January 2024
- Published by Engineering and Technology Publishing in Journal of Advances in Information Technology
- Vol. 15 (1) , 1-9
A Machine Learning Approach for Stroke Differential Diagnosis by Blood Biomarkers
- 1 January 2024
- Published by Engineering and Technology Publishing in Journal of Advances in Information Technology
- Vol. 15 (1) , 1-9
Open AccessA Machine Learning Approach for Stroke Differential Diagnosis by Blood Biomarkers
- Open Access
A plasma peptidomic signature reveals extracellular matrix remodeling and predicts prognosis in alcohol-related hepatitis
Alcohol-related hepatitis (AH) is plagued with high mortality and difficulty in identifying at-risk patients. The extracellular matrix undergoes significant remodeling during inflammatory liver injury that can be detected in biological fluids and potentially used for mortality prediction. EDTA plasma samples were collected from AH patients (n= 62); Model for End-Stage Liver Disease (MELD) score defined AH severity as moderate (12-20; n=28) and severe (>20; n=34). The peptidome data was collected by high resolution, high mass accuracy UPLC-MS. Univariate and multivariate analyses identified differentially abundant peptides, which were used for Gene Ontology, parent protein matrisomal composition and protease involvement. Machine learning methods were used on patient-specific peptidome and clinical data to develop mortality predictors. Analysis of plasma peptides from AH patients and healthy controls identified over 1,600 significant peptide features corresponding to 130 proteins. These were enriched for ECM fragments in AH samples, likely related to turnover of hepatic-derived proteins. Analysis of moderate versus severe AH peptidomes showed a shift in abundance of peptides from collagen 1A1 and fibrinogen A proteins. The dominant proteases for the AH peptidome spectrum appear to be CAPN1 and MMP12. Increase in hepatic expression of these proteases was orthogonally-validated in RNA-seq data of livers from AH patients. Causal graphical modeling identified four peptides directly linked to 90-day mortality in >90% of the learned graphs. These peptides improved the accuracy of mortality prediction over MELD score and were used to create a clinically applicable mortality prediction assay. A signature based on plasma peptidome is a novel, non-invasive method for prognosis stratification in AH patients. Our results could also lead to new mechanistic and/or surrogate biomarkers to identify new AH mechanisms. Lay summary: We used degraded proteins found the blood of alcohol-related hepatitis patients to identify new potential mechanisms of injury and to predict 90 day mortality.
Open AccessA plasma peptidomic signature reveals extracellular matrix remodeling and predicts prognosis in alcohol-related hepatitis
Alcohol-related hepatitis (AH) is plagued with high mortality and difficulty in identifying at-risk patients. The extracellular matrix undergoes significant remodeling during inflammatory liver injury that can be detected in biological fluids and potentially used for mortality prediction. EDTA plasma samples were collected from AH patients (n= 62); Model for End-Stage Liver Disease (MELD) score defined AH severity as moderate (12-20; n=28) and severe (>20; n=34). The peptidome data was collected by high resolution, high mass accuracy UPLC-MS. Univariate and multivariate analyses identified differentially abundant peptides, which were used for Gene Ontology, parent protein matrisomal composition and protease involvement. Machine learning methods were used on patient-specific peptidome and clinical data to develop mortality predictors. Analysis of plasma peptides from AH patients and healthy controls identified over 1,600 significant peptide features corresponding to 130 proteins. These were enriched for ECM fragments in AH samples, likely related to turnover of hepatic-derived proteins. Analysis of moderate versus severe AH peptidomes showed a shift in abundance of peptides from collagen 1A1 and fibrinogen A proteins. The dominant proteases for the AH peptidome spectrum appear to be CAPN1 and MMP12. Increase in hepatic expression of these proteases was orthogonally-validated in RNA-seq data of livers from AH patients. Causal graphical modeling identified four peptides directly linked to 90-day mortality in >90% of the learned graphs. These peptides improved the accuracy of mortality prediction over MELD score and were used to create a clinically applicable mortality prediction assay. A signature based on plasma peptidome is a novel, non-invasive method for prognosis stratification in AH patients. Our results could also lead to new mechanistic and/or surrogate biomarkers to identify new AH mechanisms. Lay summary: We used degraded proteins found the blood of alcohol-related hepatitis patients to identify new potential mechanisms of injury and to predict 90 day mortality.