Business Innovation

February 3, 2025

Digital pathology aims to unlock secrets hidden in tissue samples

The pilot project, which is expected to digitize 500,000 slides over three years, aims to open avenues for research, assist with access to whole-slide images for education and clinical meetings (of tumor boards, for example), and facilitate planning for the implementation of digital pathology for diagnostic purposes. 

At Vanderbilt Medical Laboratories, a transformation in the practice of pathology is underway. Traditional glass slides that are used for diagnostic purposes are being converted to high-resolution digital images — a shift to what’s called “digital pathology.” 

“This shift from traditional glass slides to digital images will improve accessibility and unlock the potential for advanced technology like artificial intelligence to enhance efficiency, accuracy and eventually patient care,” said Fedaa Najdawi, MD, medical director of Informatics and Digital Pathology at VUMC.  

Fedaa Najdawi, MD, is leading the effort to implement digital pathology at VUMC. (photo by Donn Jones)

Najdawi, a gastrointestinal pathologist who is leading the implementation of digital pathology, joined VUMC last year from the company PathAI, where she developed AI-driven models to score disease activity and assess other tissue characteristics in inflammatory bowel disease samples. 

“AI can assess changes within the tissue beyond what the human pathologist’s eye can perceive,” said Najdawi, assistant professor of Pathology, Microbiology and Immunology. “There’s a lot of information locked in the tissue, and we’re hopefully going to get access to it. The first step is to scan the slides.” 

To take the first step, VUMC has partnered with Pramana Inc., an AI-enabled health tech company, to use its autonomous whole-slide imaging system. 

This system, possessing significant capabilities, is located in a specialized room at Vanderbilt Medical Laboratories. It features four scanning compartments that surround a central platform with a robotic arm. The robotic arm twists and turns like a swan’s long neck, moving glass slides from their starting basket positions into the scanners and then out of them when scanning is complete. Each slide is scanned at multiple magnifications, and the scan time varies based on the tissue characteristics.

The Pramana Inc. scanner has four scanning compartments that surround a central platform with a robotic arm. (photo by Donn Jones)

“The Pramana scanner has automated quality control and AI-driven precision scanning, which really sets high standards for reliability as well as image quality, and I believe this will help us drive innovation in pathology,” Najdawi said. 

In a pilot project that started in early January, VUMC is scanning about 900 slides per day that include tissues from cancer, liver and heart biopsies. The tissue sample slides selected for scanning are prepared in the laboratory and examined by pathologists according to the standard clinical diagnostic protocols. After they have been used for diagnosis, they are routed for digitization. 

The robotic arm moves glass slides from their starting basket positions into the scanners and then out of them when scanning is complete. (photo by Donn Jones)

The pilot project, which is expected to digitize 500,000 slides over three years, aims to open avenues for research, assist with access to whole-slide images for education and clinical meetings (of tumor boards, for example), and facilitate planning for the implementation of digital pathology for diagnostic purposes. 

“We are building an inventory of whole-slide images that will pave the way for collaboration and will empower our researchers to extract valuable insights from tissue samples. For example, by integrating AI for biomarker discovery, we are helping advance the mission of patient-centered precision medicine,” Najdawi said. “Within the next six to 12 months, we anticipate having a sufficient inventory to launch research projects across the institution.” 

Nearly half of academic medical centers in the United States have adopted whole-slide imaging for digital pathology, but only about 10% have implemented digital pathology for routine primary diagnosis, she noted. 

Each scanning compartment contains a microscope that can scan slides at multiple magnifications. (photo by Donn Jones)

Najdawi, while recognizing the difficulty of the transition, is confident in Vanderbilt’s ability to implement digital pathology for clinical diagnoses. 

“My experience spans both academia and industry, giving me a broad understanding of the digital pathology landscape and its various vendors,” Najdawi said. “Having that combination of experiences has helped me know what questions to ask and what to assess in terms of the infrastructure and technical system requirements. The pilot project — in addition to being valuable for research and education — will help us know what we need to go forward toward digital diagnostics.” 

“Digital pathology has been an important focus of our strategic plan,” said Alice Coogan, MD, the Dorothy Beryl and Theodore R. Austin Professor of Pathology and chair of the Department of Pathology, Microbiology and Immunology. “This partnership with Pramana Inc. and its application for research innovation and education is an exciting first step in our journey toward using digital pathology for clinical diagnosis. 

“The recruitment of Dr. Najdawi, with her strong background in digital pathology and AI, was instrumental in getting this project off the ground. Her leadership in this area has been transformational.”

Najdawi completed her gastrointestinal and liver pathology fellowship at Brigham and Women’s Hospital/Harvard Medical School and trained in anatomical pathology in the United States and Australia. She is board-certified by the American Board of Pathology, and she is a Fellow of the College of American Pathologists. 

Before joining VUMC, Najdawi was the director of Pathology at PathAI, leading AI-powered diagnostic innovations. She has multiple publications on precision medicine and AI applications in pathology and holds a patent in computational pathology related to her work on AI-powered histological assessment for inflammatory bowel disease. She serves on the Digital Pathology Association Creative Content Committee and the USCAP (United States and Canadian Academy of Pathology) Finance Committee, and she was recognized as an ASCP (American Society for Clinical Pathology) 40 Under 40 honoree.