Updates Research Publications Experience & Education Teaching Service & Awards Skills Team Contact
Tahir Mahmood

Tahir Mahmood

Assistant Professor
Division of Electronics & Electrical Engineering, Dongguk University

Can an AI model be trusted once it leaves the lab?

Uncertainty and explainability are the trust axis; edge, low-resource, and challenging environments are the deployment axis. The intersection stays mostly empty, because uncertainty is studied on well-resourced models and efficiency is measured by accuracy alone.

That intersection is my research: trustworthy AI under deployment constraints. I build systems that stay calibrated and explainable on the hardware and in the conditions where they actually run, with applications in medical imaging and aerial and drone imagery.

Recent Updates

2025
Paper"A lightweight hierarchical feature fusion network for surgical instrument segmentation" accepted in Information Fusion (IF 15.5).
2025
ProjectNew project launched on Physics-Informed Neural Networks for coupled lithium diffusion and stress simulation in battery particles.
Aug 2024
AwardReceived the Best Lecture award from the ACM Student Chapter at the 2024 ACM Internal Level Refresher Course, SRM Institute of Science and Technology, India.
2024
Paper"A novel convolution transformer-based network for histopathology-image classification" accepted in Engineering Applications of Artificial Intelligence.
2023 – Present
ServiceTeam Member, BK21 FOUR (Next-Generation Mobile and Applied Technology Expert Training Team), funded by the National Research Foundation of Korea.

Research

Trustworthy AI under deployment constraints, applied to medical imaging and aerial and drone imagery.

Axis 1 · Trust
Uncertainty & Explainability

Whether a model knows when it is wrong, and whether it can say why. Calibration and conformal prediction, separating epistemic from aleatoric uncertainty, out-of-distribution detection, and selective prediction that defers to a human rather than guessing. On the explainability side, faithful explanation: concept-based and language-grounded reasoning that can be checked against the evidence, rather than post-hoc saliency that merely looks convincing.

Calibration Conformal Prediction OOD Detection Selective Prediction Faithful Explanation
Axis 2 · Deployment
Edge AI & Challenging Environments

Whether a model works at all where it is needed. Efficient architectures for surgical carts, endoscopy systems, onboard drone processors, and consumer devices, where on-device inference also keeps sensitive data from leaving the site. Alongside this, robustness to the conditions real data arrives in: scanner and stain variation, motion blur, occlusion, smoke, altitude and viewpoint change, and scenes never represented in training.

Edge AI Model Compression Low-Resource Settings Domain Generalization On-Device Privacy
The intersection is where the open problems are. Uncertainty is studied on well-resourced models, and efficiency is studied while measuring only accuracy. But compression degrades trust disproportionately: quantization and distillation damage calibration, and pruned models often fail hardest on the cases that were already underrepresented. Calibrated uncertainty inside a model small enough to run on a clinical device, or faithful explanations that survive quantization, remain open, and that is where my group works.
Application Areas
Computational Pathology Surgical Instrument Segmentation Endoscopic & Colorectal Screening Aerial & Drone Imagery Urban Safety Monitoring Precision Agriculture Agentic AI Physics-Informed Neural Networks
Current Research Projects
🔬
Agentic AI · Consumer Electronics
Agentic AI for Autonomous and Interpretable Colorectal Polyp Segmentation

A consumer-electronics-oriented agentic AI framework for polyp segmentation and clinical decision support, designed to operate across hospital workstations, mobile devices, and patient-facing consumer electronics.

🧬
Segmentation · Multimodal LLM
Adaptive Edge and Context Fusion with Language-Informed Diagnostic Reasoning

EMCAF-Net, an Edge-Aware Multi-Scale Contextual Attention Fusion Network for accurate cell nuclei segmentation in cancer diagnosis, integrating edge-aware multi-scale fusion for boundary localization with a multimodal LLM for automated diagnostic reasoning.

🔋
Physics-Informed ML
PINNs for Coupled Lithium Diffusion and Stress Simulation in Battery Particles

A physics-informed neural network framework for the coupled chemo-mechanical PDEs governing lithium transport and stress evolution in battery electrode particles, replacing traditional finite element methods. Anisotropic, SOC-dependent diffusivity tensors derived from molecular dynamics simulations are embedded as inputs, enabling accurate modelling of irregular particle geometries from X-ray tomography data.

In collaboration with Dr. Ijaz

Publications

30
Total
10
First Author
20
Co-author
2025
1
Tahir Mahmood, Ganbayar Batchuluun, Seung Gu Kim, Jung Soo Kim, Kang Ryoung Park
"A lightweight hierarchical feature fusion network for surgical instrument segmentation in internet of medical things."
Information Fusion, 2025.IF 15.5
Link ↗
2
Naqqash Dilshad, Sana Parez, Samee Ullah Khan, Mustaqeem Khan, Tahir Mahmood, Norah Saleh Alghamdi, JaeSeung Song
"YOLOFlame: Benchmarking vehicle fire and smoke detection in urban environments."
Alexandria Engineering Journal, 2025.IF 6.8
3
Faiqa Maqsood, Zhenfei Wang, Muhammad Mumtaz Ali, Baozhi Qiu, Tahir Mahmood, Raheem Sarwar
"An efficient enhanced feature framework for grading of renal cell carcinoma using histopathological images."
Applied Intelligence, 2025.
2024
1
Tahir Mahmood, Zeeshan Ullah, Atif Latif, Binish Arif Sultan, et al.
"Computer-Aided Diagnosis in Spontaneous Abortion: A Histopathology Dataset and Benchmark for Products of Conception."
Diagnostics, 2024.
Link ↗
2
Tahir Mahmood, Abdul Wahid, Jin Seong Hong, Seung Gu Kim, Kang Ryoung Park
"A novel convolution transformer-based network for histopathology-image classification using adaptive convolution and dynamic attention."
Engineering Applications of Artificial Intelligence, 2024.IF 8.0
3
Abdul Wahid, Tahir Mahmood, Jin Seong Hong, Seung Gu Kim, et al.
"Multi-path residual attention network for cancer diagnosis robust to limited hyperspectral pathological training data."
Engineering Applications of Artificial Intelligence, 2024.IF 8.0
4
Muhammad Zubair, Muhammad Owais, Tahir Mahmood, Saeed Iqbal, Syed Muhammad Usman, Irfan Hussain
"Enhanced gastric cancer classification and quantification interpretable framework using digital histopathology images."
Scientific Reports, 2024.IF 3.9
2023
1
Tahir Mahmood, Seung Gu Kim, Se Hyun Nam, Kang Ryoung Park
"CFFR-Net: A channel-wise features fusion and recalibration network for surgical instruments segmentation."
Engineering Applications of Artificial Intelligence, 2023.IF 8.0
2
Nadeem Ullah, Tahir Mahmood, Seung Gu Kim, Se Hyun Nam, Haseeb Sultan, Kang Ryoung Park
"DCDA-Net: Dual-convolutional dual-attention network for obstructive sleep apnea diagnosis from single-lead electrocardiograms."
Engineering Applications of Artificial Intelligence, 2023.IF 7.5
3
Tahir Mahmood, Jiho Choi, Kang Ryoung Park
"Artificial intelligence-based classification of pollen grains using attention-guided pollen features aggregation network."
Journal of King Saud University – Computer and Information Sciences, 2023.
2022
1
Tahir Mahmood, Se Woon Cho, Kang Ryoung Park
"DSRD-Net: Dual-stream residual dense network for semantic segmentation of instruments in robot-assisted surgery."
Expert Systems with Applications, 2022.IF 7.5
2
Tahir Mahmood, Seung Gu Kim, Ja Hyung Koo, Kang Ryoung Park
"Artificial intelligence-based tissue phenotyping in colorectal cancer histopathology using visual and semantic features aggregation."
Mathematics, 2022.
2021
1
Tahir Mahmood, Muhammad Owais, Kyoung Jun Noh, et al.
"Accurate segmentation of nuclear regions with multi-organ histopathology images using artificial intelligence."
Journal of Personalized Medicine, 2021.
2020
1
Tahir Mahmood, Muhammad Arsalan, Muhammad Owais, Min Beom Lee, Kang Ryoung Park
"Artificial intelligence-based mitosis detection in breast cancer histopathology images."
Journal of Clinical Medicine, 2020.

Full list on Google Scholar ↗

Experience & Education

Mar 2023 – Present
Assistant Professor
Division of Electronics and Electrical Engineering, Dongguk University, Seoul, South Korea
  • Teach graduate-level courses in Artificial Intelligence and Robot Vision, emphasizing agentic AI systems, real-world deployment, and challenge-based learning aligned with frontier AI research.
  • Supervise and mentor Master's and Ph.D. students in medical image analysis, precision agriculture, and agentic AI systems.
  • Team Member, BK21 FOUR — Brain Korea 21 (Next-Generation Mobile and Applied Technology Expert Training Team), funded by the National Research Foundation of Korea (NRF).
  • Thesis Evaluation Committee Member, Division of Electronics and Electrical Engineering.
Oct 2017 – Feb 2019
Machine Learning Engineer
Air-Fi, Barcelona, Spain
  • Contributed to the development and deployment of the AIR-FI Indoor Positioning System.
  • Analyzed data and gathered client requirements for system customization.
  • Built a web-based platform for system visualization and management using React/Node.js.
  • Conducted performance testing and troubleshooting, reducing latency in data processing.
Apr 2016 – Sep 2017
Lecturer
Superior University, Lahore, Pakistan
  • Delivered undergraduate courses in Computer Science and Software Engineering with a focus on practical, industry-relevant skills.
  • Contributed to curriculum development aligned with emerging technology trends.
  • Organized technical workshops and seminars; served on university academic committees.
  • Courses taught: Programming Fundamentals, Artificial Intelligence, Software Requirements Engineering, Project Management.
Education
Mar 2019 – Feb 2023
Doctor of Engineering
Division of Electronics and Electrical Engineering, Dongguk University, Seoul, South Korea
Major: Multimedia Contents and Signal Processing · CGPA 3.8/4.0
Dissertation: "A study on deep learning-based classification, detection, and segmentation for histopathology images."
2013 – 2015
M.S. in Computer Science
Department of Computer Science, COMSATS University, Islamabad, Pakistan
CGPA 3.5/4.0
Dissertation: "An automated mitosis recognition technique for breast cancer histopathology slide images."
2009 – 2013
B.S. in Computer Science
Department of Computer Science, COMSATS University, Islamabad, Pakistan
CGPA 3.1/4.0
Dissertation: "Intelligent warehouse management system."

Teaching

2023 – Present Graduate
Artificial Intelligence
Dongguk University
Graduate-level AI with an emphasis on agentic AI systems, real-world deployment, and challenge-based learning aligned with frontier AI research.
2023 – Present Graduate
Robot Vision
Dongguk University
Visual perception for robotic systems, covering representation, detection and segmentation, and the deployment of vision pipelines in real environments.
2016 – 2017 Undergraduate
Programming Fundamentals · Artificial Intelligence · Software Requirements Engineering · Project Management
Superior University, Lahore
Undergraduate courses in Computer Science and Software Engineering, focused on practical, industry-relevant skills.
Student supervision. Currently supervising 2 Ph.D. and 3 Master's students working on medical image analysis, precision agriculture, and agentic AI systems. Also serving as a Thesis Evaluation Committee Member in the Division of Electronics and Electrical Engineering.

Academic Service & Awards

Editorial Roles
  • Editorial Board Member — Advanced Multidisciplinary Engineering Journal (AMEJ)
  • Guest Editor — Artificial Intelligence in Medical Image Analysis and Clinical Decision Support, Diagnostics, MDPI
  • Guest Editor — Next-Generation Medical Imaging AI: Deep Learning, Computer Vision with Multi-Task and Continual Learning, AI, MDPI
  • Guest Editor — Latest News in Digital Pathology, Diagnostics, MDPI
Peer Reviewer
  • Expert Systems with Applications (Elsevier)
  • Engineering Applications of Artificial Intelligence (Elsevier)
  • Biomedical Signal Processing and Control (Elsevier)
  • Computers and Electrical Engineering (Elsevier)
  • IEEE Transactions on Consumer Electronics (IEEE)
  • Diagnostics, Cancers, Bioengineering (MDPI)
Honors & Awards
Best Lecture Award — 2024
Awarded by the ACM Student Chapter at the 2024 ACM Internal Level Refresher Course, School of Computing, SRM Institute of Science and Technology, India (Aug 12–16, 2024).
Research Stay — Pompeu Fabra University
Invited by the Physense research group for a two-month research stay in Barcelona, Spain.
SRD Scholarship (approx. $50,000) — 2019–2023
Awarded by Dongguk University, Seoul, Republic of Korea, for doctoral studies.
ICT Scholarship (approx. $100,000) — 2009–2013
Awarded by National ICT R&D, Pakistan, for the Bachelor's degree in Computer Science.
Honorary Reporter — 2021
Korean Culture and Information Service, Seoul, Republic of Korea (May 20, 2021).
Vice President — 2022–2023
Volunteer, Pakistan Students and Scholars Association (April 2022 – August 2023).
Graduate Committee Member
Evaluation and guidance for Master's and Ph.D. candidates.

Technical Skills

Programming & Development
ExpertMATLAB — image processing & computer vision toolboxes
ExpertPython — OpenCV, NumPy, scikit-learn, Pandas, Matplotlib, Tkinter
ExpertCUDA — installation and GPU processing (Windows & Linux)
IntermediateJava — GUI, file handling
IntermediateC, C#, C++
Deep Learning Frameworks
ExpertPyTorch — tensors, data loaders, neural networks, transformers
ExpertTensorFlow & Keras — tensors, data loaders, neural networks, transformers
ExpertMATLAB Deep Learning Toolbox
LLMs & Agentic AI
ExpertAgentic AI frameworks — multi-agent pipelines, autonomous clinical decision support
IntermediateLangChain — agent orchestration, tool use, RAG pipelines
IntermediatePrompt engineering & LLM fine-tuning — LoRA, instruction tuning
Operating Systems
ExpertUbuntu / Unix
ExpertMicrosoft Windows

Team & Collaborations

Host Professor & Research Advisor
KP
Dr. Kang Ryoung Park
Professor, Dongguk University
Computer Vision, Biometrics
Long-term research collaborator in computer vision and biometric systems, Seoul, Korea.
Research Collaborators
TZ
Dr. Talat Zehra
Associate Professor
Liaquat National Hospital and Medical College, Karachi
Computational pathology, including mitosis detection in leiomyosarcoma and ongoing work on automatic chorionic villi recognition.
NJ
Dr. Nazish Jaffar
Associate Professor
Jinnah Sindh Medical University, Karachi
AI-driven histopathology analysis, focusing on breast and colorectal cancer diagnosis.
ND
Dr. Naqqash Dilshad
Assistant Professor
Sejong University, Seoul
Vision-based urban safety systems, including vehicle fire and smoke detection in real-world environments.
NA
Dr. Najah Alsubaie
Assistant Professor
Princess Nourah bint Abdulrahman University, Saudi Arabia
Histopathology image analysis; jointly submitted a national-level research grant proposal.
SA
Dr. Shahzad Ahmed
Assistant Professor
Hanyang University, Seoul
AI-based vital sign monitoring using ultra-wideband radar and multimodal sensing technologies.
Graduate Students
My group currently includes 2 Ph.D. and 3 Master's students working across computational pathology, medical image segmentation, precision agriculture, and agentic AI systems. Individual student profiles will be listed here soon.

Contact

Affiliation

Division of Electronics and Electrical Engineering
Dongguk University
Seoul
South Korea

Prospective Students

I am looking for motivated graduate students interested in computational pathology, surgical AI, agentic AI systems, and physics-informed machine learning. If you are interested in joining my group, please send your CV and a brief research statement to my email.