Md Shakhawat Hossain, PhD
Associate Professor, Kochi University of Technology (KUT)
Academic Achievements
Research impact and career highlights
52
Publications
861
Citations
10
Awards & Grants
32
Students Supervised
Research Impact
Metrics highlighting research contributions
16
H-Index
6
International Collaborations
8
Invited Talks
Work Experience
Professional career and academic positions
Associate Professor
Kochi University of Technology , Japan
School of Informatics
Currently serving as an Associate Professor at Kochi University of Technology, I lead the Medical Imaging and AI Laboratory (AIM), also known as the Hossain Lab, as the principal investigator. In addition, I am a member of the Doctoral Program …
Assistant Professor
Independent University, Bangladesh , Bangladesh
Computer Science and Engineering
Previously worked as an Assistant Professor at Independent University, Bangladesh (IUB), where I led the Medical Imaging and AI Research Group. I also served as the Coordinator of the Resource and Lab Development Committee, a member of the Outcome-Based …
Senior Consultant
AQ Chowdhury Science & Synergy , Bangladesh
Electron microscopy
Worked as a Senior Consultant to establish a clinical-grade electron microscopy system in Government hospitals in Bangladesh.
Head of Medical Imaging and AI Research Group
RIoT Research Center, IUB , Bangladesh
Computer Science and Engineering
Served as the Head of the Medical Imaging and AI Research Group at the RIoT Research Center, IUB. Secured BDT 3,000,000 in research funding, including two government grants. Published over 20 peer-reviewed papers, established four international and four national collaborations, …
Senior Research Engineer (Machine Learning in Medical Imaging)
University of Oxford , UK
Department of Engineering Science
Received an Offer from Department of Engineering Science, Oxford University to work as an Senior Research Engineer (Machine Learning in Medical Imaging). However, I could not continue this job and had to return to my home country due to a …
Assistant Professor
American International University-Bangladesh , Bangladesh
Computer Science
Worked as an Assistant Professor of Computer Science, supervising over 30 master's and bachelor's students. Many of them are now pursuing doctoral or master's degrees in countries including the United States, Japan, France, and Malaysia.
Research Fellow
Memorial Sloan Kettering Cancer Center, USA , USA
Department of Pathology
Worked as a Research Fellow at Memorial Sloan Kettering Cancer Center under the Warren Alpert Fellowship. Contributed to a project focused on developing a multimodal database for cancer diagnosis.
Graduate Research Assistant
Memorial Sloan Kettering Cancer Center, USA , USA
Department of Pathology
Worked as a Graduate Researcher on a project to design and develop a decision support system for providing targeted therapy (Trastuzumab) to breast cancer patients. The system utilized bright-field microscopy as an alternative to the standard fluorescence microscopy used …
Visiting Researcher
Memorial Sloan Kettering Cancer Center, USA , USA
Department of Pathology
Contributed to a project focused on designing and developing a decision support system for administering targeted therapy (Trastuzumab) to breast cancer patients. The system employed bright-field microscopy as an alternative to fluorescence microscopy, which is commonly used in current …
Lecturer
ZH Sikder University of Science and Technology , Bangladesh
Computer Sceince and Engineering
Worked as a Lecturer of Computer Science
Software Developer
RajIT Software Solutions Ltd , Bangladesh
Developed software for academic institutions using PhP (CodeIgniter), JavaScript, and MySQL.
Education
Academic background and qualifications
Ph.D. in Information and Communications Engineering
Tokyo Institute of Technology , Japan
Medical Imaging
Dissertation: Quality evaluation of whole slide images for the purpose of automated analysis and diagnosis in digital pathology
This thesis focuses on developing a quality evaluation method for whole slide images (WSIs) to ensure their suitability for automated analysis in digital pathology. While WSIs offer significant potential for …
Master of Engineering in Information Processing
Tokyo Institute of Technology , Japan
Medical Imaging
Dissertation: Artifact Dectection from Whole Slide Images
Accurate artifact detection in Whole Slide Images (WSIs) is crucial for ensuring reliable automated analysis in digital pathology. Artifacts such as tissue folds, air bubbles, focus blur, and scanner …
Master of Science in Computer Science and Engineering
University of Rajshahi , Bangladesh
Image Processing
Dissertation: Multiband approach for Image Quality Enhancement using Wavelet
This thesis presents a multiband approach using wavelet transform to enhance image quality for computer-aided diagnosis. By effectively reducing noise and improving image clarity, the method supports more accurate …
BSc in Computer Science and Engineering
University of Rajshahi , Bangladesh
Image Processing
Dissertation: An Interactive Web-based Image Processing System
This thesis presents the development of an interactive web-based image processing system built using JavaScript. The system enables users to perform various image editing and processing operations online without …
Awards & Grants
Recognition and funding achievements
Multimodal AI for Breast Cancer Subtype Analysis and Precision Therapy Recommendation
Kochi University of Technology, Japan
This project aims to develop a multimodal AI system that analyzes diverse breast cancer data—including histopathology, clinical, genetic, and imaging data—to accurately determine cancer subtypes and recommend personalized treatment. The system is designed to improve diagnostic precision and reduce reliance on expensive, expert-driven methods. By integrating all relevant data sources, the AI model will provide a low-cost, scalable solution that supports precision medicine and expands access to high-quality breast cancer care.
AI-guided Endoscope Vision for Real-Time Navigation, Examination and Surgery to Enhance Endoscopists’ Efficiency and Improve Patients’ Comfort
Ministry of Science and Technology, Bangladesh
Digestive diseases, ranging from harmless lesions to life-threatening cancer, are highly prevalent worldwide and have increased rapidly in the last 20 years at a rate of 67.87%. Around 40 million Americans suffer from digestive diseases, which result in millions of endoscopy and cost billions each year. Similarly, Bangladesh's digestive diseases and endoscopy burden has been increasing. Traditionally, an endoscopist performs the endoscopy and navigates the endoscope, a thin, flexible tube with a camera, light, water jet and surgical instrument within the narrow digestive tract. The endoscopist observes the endoscope-captured video on a screen in real time to navigate the endoscope, examine anomalies and perform surgery. He also flashes water and uses an instrument for the surgery. This manual process demands a high level of physical and mental effort and has three significant issues: 1) it causes discomfit and injury to the patient due to inappropriate movement within the delicate tissues …
Tumor Cellularity Assessment of Breast Cancer after Neoadjuvant Chemotherapy
Independent University, Bangladesh Vice Chancellor Research Grant
This research project focuses on developing an automated method to assess residual tumor cellularity (RTC) in breast cancer patients after chemotherapy. Currently, RTC evaluation is performed manually by expert pathologists, a process that is time-consuming, labor-intensive, and prone to variability between observers. This manual approach also contributes to delays in diagnosis, particularly in regions like Bangladesh, where pathologist availability is limited. The proposed AI-based system aims to replicate the expertise of pathologists, reducing their workload and eliminating variability. Designed to deliver assessments in under a minute, this method significantly accelerates diagnosis and improves patient care. By using digital camera-mounted microscopes, which are common in most hospitals, the system offers a cost-effective and practical solution for faster and more accurate breast cancer diagnosis.
Predicting the Therapeutic Response of Bevacizumab Humanized Monoclonal Therapy on Ovarian Cancer Patients
Independent University, Bangladesh Sponsored Research Grant
Ovarian cancer (OC) is the leading cause of death among gynecologic cancers worldwide. In Dhaka, a recent survey of 117 cancer patients found that 8.1% of women suffer from OC, and this rate is expected to double within the next 20 years. The standard treatment involves surgical removal of cancerous tissues followed by chemotherapy, but this approach often results in high tumor recurrence, progression, and low survival rates. Recently, the FDA approved Bevacizumab therapy for ovarian cancer, which can shrink tumors. However, some patients experience serious side effects such as hypertension (10%), pulmonary hemorrhage (6.5%), intestinal perforation (3.5%), intracranial bleeding (2.0%), cardiac toxicity, and delayed wound healing if administered inappropriately. Additionally, the treatment is expensive. This research aims to develop an automated system to predict how OC patients will respond to Bevacizumab, improving personalized treatment and patient safety.
Low-cost and Rapid HER2 Grading of Breast Cancer Patients in Bangladesh
Ministry of Education, Bangladesh
Each year, over 13,000 women in Bangladesh are diagnosed with breast cancer, and about 7,000 die. Late diagnosis and high costs are major challenges. HER2 protein helps assess cancer severity and guide treatment. Currently, Hematoxylin-Eosin (H&E) staining is used to identify cancer areas, but grading HER2 requires expensive and slow tests like Immunohistochemistry (IHC) and Fluorescence In Situ Hybridization (FISH). Many patients cannot afford these tests, especially since over 20% of the population lives below the poverty line. This research proposes using artificial intelligence to predict HER2 grades directly from affordable and quick H&E images. This AI-based approach aims to offer a low-cost, fast, and accessible method for HER2 grading, improving breast cancer diagnosis and treatment in Bangladesh.
Machine learning assisted decision (MAD) Support System for the Diagnosis and Treatment Planning of Breast Cancer Patients
Independent University, Bangladesh Sponsored Research Grant
In Bangladesh, over 13,000 women are diagnosed with breast cancer (BC) annually, and more than 7,000 die—resulting in a death rate of 53.8%. Planning treatment for BC is complex, requiring consideration of factors like age, tumor grade, size, gene amplification, heart condition, and menopausal status. Access to such comprehensive care often comes at high out-of-pocket expenses, which is unaffordable for many, especially as 21.8% of the population lives in poverty. Diagnosis typically begins with a Hematoxylin and Eosin (H&E) stain analysis, followed by molecular tests such as Immunohistochemistry (IHC) and Fluorescence In Situ Hybridization (FISH). While H&E-based tests are fast (within a few hours) and low-cost (under 3000 BDT), IHC and FISH cost between 11,000–20,000 BDT and take 5–7 days. Therefore, the current diagnostic process is expensive and time-consuming. This research aims to develop a fast, affordable, and automated diagnosis system to improve breast cancer care.
Practical Image Quality Evaluation for Whole Slide Image Analysis
MURATA Science Foundation Research Grant, Japan
Pathology traditionally relies on microscopes to examine tissue samples, but this method is time-consuming, difficult to share, and degrades over time. Whole Slide Imaging (WSI) scanners offer a digital alternative by converting slides into high-resolution images. However, poor image quality—due to focus errors or noise—can lead to diagnostic errors. Our research developed an automated method to evaluate image quality and recommend rescanning if needed. The goal is to integrate this system into hospitals, ensuring WSI is practical and reliable for routine pathology.
Professional Activities
Editorial boards, reviewing, and service
Editorial Boards
5+
Serving on editorial boards of leading journals in AI and medical imaging
Peer Review
10
Regular reviewer for top-tier conferences and journals in the field
Committee Service
10+
Active member of program committees for major conferences
Research Impact Metrics
Comprehensive view of research contributions
52
Publications
861
Citations
16
H-Index
80
Students
7
Grants
5+
Patents