Among AI’s great promises in relation to medicine is its potential to use existing patient data—including MRIs—to identify and diagnose potential problems. Doing so has many potential benefits, including lower costs and fewer invasive patient procedures.
Among the researchers making good on this promising AI potential is Mohamed Shehata.
Shehata is a postdoctoral fellow in the Department of Bioengineering at the University of Louisville. He’s won numerous awards for his work using machine learning to design computer-aided diagnostic and prediction systems for the early detection of medical issues, including tumors and post-transplant kidney rejections.
Shehata is also one of Computing’s Top 30 Early Career Professionals for 2024. In the following Q&A, he describes
- His award-winning dissertation on how machine learning models could be applied to patent data and medical imaging for early diagnosis of kidney diseases
- The non-invasive, cost-effective AI-based system he developed to detect acute renal transplant rejection at earlier stages than were previously possible
- How improving patient outcomes and pushing the boundaries of medical science drive his work and fuel his curiosity
- What he’s learned lately, including about the value of failure and of collaborating with experts in other fields
You received the MIT Technology Review Award for Innovators Under 35 in the MENA region in 2021. Can you share the project or innovation that led to this recognition, and what impact it has had on the field of medical imaging?
In 2021, I received this MIT Technology Review Award for developing an AI-based system for the early detection of acute renal transplant rejection using MRIs. This system leverages deep learning algorithms to analyze MRI scans, identifying subtle changes in kidney tissue that may indicate rejection at an early stage.
The innovation improves diagnostic accuracy, enabling earlier intervention and better outcomes for transplant patients. Automating image analysis reduces human error and enhances the efficiency of medical professionals. This system is non-invasive, cost-effective, and rapid, helping make advanced diagnostics more widely available. It has contributed to advancing medical imaging by integrating AI into real-time clinical decision-making processes.
Your dissertation was awarded the John M. Houchens Prize for the best “and most meritorious” dissertation at the University of Louisville in Fall 2022. What were the key findings of your dissertation on the role of machine learning in the early diagnosis of kidney diseases, and how do they contribute to the field?
My dissertation focused on the role of machine learning in the early diagnosis of kidney diseases, including renal rejection post-transplantation and renal cancer. The key findings demonstrated that machine learning models, when applied to patient data and medical images, could identify early biomarkers of kidney diseases with higher accuracy than traditional methods.
I developed predictive models that analyzed factors like clinical biomarkers and imaging data, achieving significant improvements in early detection. These findings contribute to the field by providing a more reliable, cost-effective, and efficient tool for diagnosing kidney diseases at earlier, more treatable stages. My research also highlighted the potential of AI in personalized treatment plans, allowing for tailored therapies. This work pushes the boundaries of medical diagnostics, especially in resource-limited settings.
You have won the R!L competition at both the postdoctoral level in 2024 and the doctoral engineering level in 2021. Can you discuss the projects you worked on for these competitions and the innovative solutions you developed?
In 2024, I won the R!L competition at the postdoctoral level with an AI-based system designed for renal cancer assessment. This system utilizes advanced machine learning algorithms to analyze medical images, improving the accuracy and speed of early cancer differentiation between malignant and benign tumors as well as the typing of malignant ones, aiding in personalized treatment plans.
In 2021, at the doctoral engineering level, I developed an AI-based system for the early detection of renal allograft rejection. By analyzing MRI scans and clinical data, this system can detect early signs of transplant rejection, enabling timely interventions.
Both projects focused on integrating AI into medical diagnostics, enhancing early disease detection, and improving patient outcomes through precision medicine. These innovative solutions represent significant advancements in the field of medical imaging and healthcare.
Receiving the Doctoral Dissertation Completion Award in 2021 is a significant achievement. What were the main challenges you faced during your doctoral research, and how did you overcome them?
One of the main challenges I faced was integrating complex machine learning models with medical data, which often required handling incomplete or noisy datasets. To overcome this, I developed advanced preprocessing techniques to clean and standardize the data, ensuring more accurate model predictions.
Another challenge was balancing the technical complexity of AI algorithms with real-world clinical applications. In addition to teamwork and the endless effort and spirit of success, I collaborated closely with healthcare professionals to ensure the models were practical and aligned with clinical needs. Additionally, the rapid pace of technological advancements required me to stay current with evolving methods, which I managed through continuous learning and collaboration. These efforts culminated in a successful dissertation, bridging the gap between AI and healthcare.
You have been awarded the EPIC Inventor Award multiple times (2019, 2020, and 2022). Can you highlight one of your key inventions and its practical applications in the medical field?
One of my key inventions, awarded the EPIC Inventor Award in 2020, is an AI-based diagnostic tool for early-stage kidney disease detection. The system analyzes and integrates medical images with clinical data to identify subtle biomarkers of kidney dysfunction at an early stage.
This invention has practical applications in early diagnosis, enabling timely interventions and preventing disease progression. It enhances the accuracy and efficiency of clinical assessments, reducing the burden on healthcare professionals and improving patient outcomes. By automating the detection process, the tool also makes kidney disease diagnosis more accessible, particularly in resource-limited settings. This innovation represents a significant step forward in using AI to revolutionize medical diagnostics.
The Exemplary Research Scholarship Award in the Bioengineering Department at the University of Louisville is a testament to your research excellence. What drives your passion for research, and how do you stay motivated to continuously contribute to the scientific community?
My passion for research is driven by a deep desire to make a tangible impact on healthcare through innovation. I am motivated by the potential to improve patient outcomes, particularly in areas like early disease detection and personalized treatment. Solving complex problems with advanced technologies, such as AI and machine learning, fuels my curiosity and commitment to pushing the boundaries of medical science.
To stay motivated, I focus on the real-world applications of my work and the opportunity to help underserved populations. Wise leadership by my mentors Prof. Ayman El-Baz and Prof. Adel Elmaghraby, collaboration with experts in medicine, Prof. Amy Dwyer, Prof. Rosemary Ouseph, and Prof. Mohamed Abou El-Ghar, and technology also keep me engaged and inspire continuous learning. Ultimately, the prospect of contributing to meaningful advancements in healthcare drives my dedication to research and my commitment to the scientific community.
Your paper was selected as the cover article for the Medical Physics Journal in 2020. Can you discuss the main contributions of this paper and its impact on the field of medical physics?
My paper presented a novel AI-driven approach for enhancing medical imaging in the detection of early-stage renal diseases. The main contribution was developing a deep-learning-based algorithm that analyzes multimodal MR images acquired from different geographical areas and integrates them with the associated clinical data. This improves the accuracy and speed of analyzing medical scans, specifically MRIs, for early biomarkers of kidney dysfunction.
This innovation significantly reduces diagnostic errors and enhances early detection, leading to better patient outcomes. The paper’s impact on medical physics lies in demonstrating how AI can optimize image analysis, reduce reliance on manual interpretations, and support precision medicine. It opened new avenues for integrating AI into clinical practice, advancing the field of medical imaging, and fostering further research in AI-driven diagnostics.
Being recognized among the Best 39 papers at the IEEE ISBI in 2016 is a notable accomplishment. What was the focus of your paper, and what were the key findings?
My paper focused on applying machine learning for early detection of renal rejection from 3D MRIs. This paper includes several novelties, including segmentation, registration, and classification. This paper provided our preliminary results on a small data set, but the concept of automatic analysis of such diffusion MRI data was new to the field. The key findings demonstrated that our model outperformed traditional methods in diagnostic performance, identifying subtle disease markers in MRIs. This approach showed significant potential for early diagnosis and improving patient outcomes through automated, real-time analysis.
You received the IEEE SPS Travel Award in 2015. How did this opportunity influence your research and professional development, and what were the key takeaways from the conference you attended?
Receiving the IEEE SPS Travel Award provided me with the opportunity to attend a prestigious conference, significantly influencing my research and professional development. It allowed me to engage with experts in signal processing and gain insights into emerging trends and technologies. Key takeaways included new methodologies for image analysis and AI applications in healthcare, which directly informed my subsequent research. The networking opportunities also led to valuable collaborations, enhancing both my academic and professional growth.
Reflecting on your career journey, from your education at Mansoura University and the University of Louisville to your current role as a post-doctoral research fellow, what are some key lessons you have learned, and how have they shaped your approach to research and innovation?
Reflecting on my career journey, the key lessons I’ve learned include the importance of persistence, collaboration, and adaptability. My education at Mansoura University and the University of Louisville taught me to embrace interdisciplinary approaches and think critically about complex problems.
During my time as a Post-Doctoral Research Fellow, I’ve learned that innovation often comes from tackling challenges with a fresh perspective and leveraging technology in novel ways. Collaboration with experts from different fields has been crucial in developing impactful solutions. Additionally, the value of failure as a learning experience has shaped my resilience and drive. These lessons have reinforced my commitment to advancing research in healthcare and developing AI-driven solutions to improve patient outcomes.
Bio: Mohamed Shehata
Mohamed Shehata is a postdoctoral fellow in the Department of Bioengineering at the University of Louisville. He has actively been working on machine and deep learning for medical imaging analysis, specifically the design of computer-aided diagnostic and prediction systems for the early detection of medical conditions. He has authored papers in more than 80 technical peer-reviewed publications, including in IEEE Transactions on Biomedical Engineering, Scientific Reports (Nature), Medical Physics, IEEE Access, Artificial Intelligence in Medicine, Insights into Imaging, PLOS One, and British Journal of Radiology. Shehata has a BSc in computer engineering and control systems from Mansoura University, Egypt; and an MS from the Electrical and Computer Engineering Department and PhD from the Computer Science and Engineering Department at the University of Louisville.
He has received more than 26 prestigious awards, including Best Oral Presentation Award in Med-Forward Conference 2025; the John M. Houchens Prize for the Most Meritorious Dissertation at University of Louisville; the MIT Technology Review Award for Innovators Under 35 Years Old; the EPIC Innovator Award; and the Exemplary Research Scholarship Award (2019– 2023). Shehata is also a guest editor for prestigious journals, including Frontiers in Medicine, Frontiers in Oncology, and Frontiers in Neuroscience.
Dig Deeper
To learn more about Shehata and his work,
Over the next few months, Tech News will highlight different Top 30 honorees each week. For a full list, see Computing’s Top 30 Early Career Professionals for 2024.
To read more about how IEEE Computer Society supports our world and its innovative thinkers through funding, education, and activities, check out its other contributions to the computing community.
