Digital-Pathology-Today

In this episode we discuss the enterprise pathology platform and how it is so much more than image management. “It’s where work gets done and data lives,” according to Krishna Bala of Proscia. We discuss the components or steps involved in digital pathology workflow. How do we make it all seamless? What is the appropriate level of pathologist involvement in the design and implementation of such systems? How can software engineers acquire domain expertise? What does integration mean and how can we incorporate exciting new AI tools into our daily practice? We discuss the benefits of an open platform and why our guest thinks this model will ultimately win out. Krishna Bala is currently Sr. VP of Engineering at Proscia, a leading provider of digital and computational pathology solutions. In this role, he leads the development of Proscia's Concentriq enterprise pathology platform, which is used by leading life sciences organizations and top diagnostic laboratories. Krishna is an experienced healthcare technologist with more than 20 years of experience in software development, data engineering, and AI/Machine-Learning. He is also an entrepreneur with several successful exits including an IPO as CTO/Founder. Krishna has a Ph.D. in Electrical Engineering from Columbia University in New York.

Today we dive into a field that hasn't existed until very recently - ethics in AI and digital pathology. What is ethics and how is it relevant in digital pathology? We all want to do the right thing, don't we? Maybe easier said than done? Do we even have a working definition of AI that people can agree on? We discuss a framework for ethics, incorporating such principles as fairness and inclusiveness, transparency, and algorithmic bias to name a few. What are the risks of artificial intelligence for pathologists, either real or imagined? Dr. Chhavi Chauhan is Director of Scientific Outreach at the American Society for Investigative Pathology, a Program Manager for the Women in AI Accelerate & Raise Program, and currently serves on the Boards for 8 different mission-driven organizations in the spheres of scholarly publishing, digital pathology, AI Ethics, & youth education. She is a former biomedical researcher, expert scholarly communicator, and a sought-after mentor in the fields of scientific research, scholarly publishing, and AI Ethics, especially for women and minorities. She is a thought leader, a renowned international speaker, and a strong advocate for equitable and accessible healthcare. She was named in the AI Makers 150 AI & Analytics Leaders & Influencers 2021 List, the 100 Brilliant Women in AI Ethics 2022 list, the Top 100 Women of the Future 2023 List, and was listed as a Finalist for the AI in Researcher Award '23 from Women In AI.

Lisa-Jean Clifford is Chief Operating Officer and Chief Strategy Officer at Gestalt Diagnostics. We discuss parallels between pathology and radiology in their digital journeys, the steps involved in going digital, how to keep the various stakeholders engaged and enthusiastic. We posed the question: “ Have we finally made the business case for digital pathology?” Lisa-Jean tells us about the user experience for pathologists in the digital environment and how many of her clients are having a faster more streamlined experience than with glass slides. We discuss the promise of new technologies in digital pathology and how in many ways, digital pathology is a fresh start for pathology. For more than 2 decades, Lisa-Jean Clifford has been a noteworthy leader in the high-tech healthcare solutions space. Her passion for making a positive impact on the lives of patients through technology can be traced back to her tenure at McKesson and IDX, now GE Healthcare, where she served in vital business development and marketing roles and to Psyche Systems, an LIS solution provider, where she was the CEO for eleven years. Now, recognized as an industry expert, she actively participates in numerous boards including the Association of Pathology Informatics and MLO’s Editorial Advisory Board. She is widely published in many top laboratory publications and noteworthy news sources, such as Forbes, CAPToday, Medical Laboratory Observer, and Health Data Management. Also, she is a highly sought-after speaker and focuses on delivering valuable content in critical areas such as lab automation including software and interoperability, digital pathology, AI in pathology, lab informatics, oncology, and women’s health.

This episode with Dr. Ken Bloom of Nucleai, we examine the beginnings of telepathology, which led to what we now know as digital pathology in the 1980s. We trace the journey to now, focusing on seminal events in the industry, such as computational pathology, computer-assisted image analysis, the development of CPT codes, the importance of the TCPC split, particularly in the reference lab setting, We also discuss the democratization of immunohistochemistry in connection with digital pathology and touch on recent developments in the assessment of tumor heterogeneity, AI algorithms, and spatial analysis. Additionally, we explore the role of pathologists in the era of precision medicine, and the potential for integration of pathology with radiology. Dr. Kenneth J. Bloom is currently the head of pathology at Nucleai. He has more than 35 years of clinical experience in pathology, oncology, telemedicine, and bioinformatics. Dr. Bloom previously served as the Chief Medical Officer at Invicro and Ambry Genetics. He was also President and Head of Oncology and Immunotherapy for Human Longevity Inc. He played an instrumental role in designing the business strategy that grew Clarient from a startup with no revenue to its acquisition by GE. He oversaw the development of hundreds of laboratory-developed tests, including those using IHC, ISH, Flow Cytometry, and Molecular methods. Dr. Bloom spent 15 years at Rush Medical College, where he held various titles, including Director of Laboratory Operations and Chief Information Officer of the Rush Cancer Institute

We know that there are certain tasks pathologists are often not very good at or may prefer not to do, such as counting cells or estimating tumor volume. Why not free them up from these tasks and allow them to focus on other areas where they can add more value? As technology accelerates and drug discovery quickens, there are more and more studies being conducted and fewer pathologists available to participate. What if we could automate some of these processes to alleviate the burden on pathologists? At the heart of clinical research, particularly in oncology, is the patient specimen, usually in the form of a paraffin block. What if we could reimagine the way we unlock the precious DNA and RNA from the paraffin block? What if researchers could view digital images of HD sections and select ones that meet their precise needs for a study before the blocks even arrive in their lab, getting exactly what they want and saving months of effort? Jon Wetzel, COO of Trimetis, is a researcher at heart, with over 24 years of experience in the life sciences and 20 years of management in startup biotech companies. He has created several companies, most recently co-founding FoundationBio, which is now part of TriMetis. Jon has served in academia and senior roles in the industry with the University of Michigan, UCLA, Genelogic, Rosetta Inpharmatics, Asterand, and AdeptBio. Jon holds several certifications from the University of Michigan for Lean and Six Sigma and is a Certified Six Sigma Black Belt. We discuss computer-assisted pathology and the suite of software tools that Jon and the team at TriMetis have been developing to accelerate progress in the biospecimen and tissue-based research field.

What does the CRO of the future look like? In diagnostics we seem to be inundated with more platforms, more analytes more “-omics” and more pathways. How do we make sense of it all? What is the future of multi-omic analysis and what role will digital pathology play in understanding biological pathways, clinical trials, multi-omics and the development of companion diagnostics? Christopher Ung, Chief Scientific Business Officer at CellCarta, has 22 years of experience in the field of digital pathology. He has made a significant impact by leading innovative solutions in the realm of companion diagnostics. He led the development of HerceptTest, and in 2009 built and connected pathology labs in four different countries with a digital pathology system. Christopher 's passion for digital pathology stems from a belief in the potential for technology to revolutionize the way pathology is practiced. He is particularly interested in the application of machine learning and artificial intelligence in digital pathology to improve the accuracy of developing and interpreting high complexity and high-density biomarkers for clinical trials and companion diagnostics. In addition to what the CRO of the future will look like, we discuss the role of digital pathology in clinical trials, multi-omics and the development of companion diagnostics. Will digital pathology play a supportive role, or will it be the centerpiece?

What is the state of the industry in 2023 in terms of driving towards unified file formats? Is middleware or software to bridge between various components and file formats still necessary and if so for how long? Why do we even have different file formats? Is it a feature or a bug? Rudy Hovelinck is the CEO of Pathomation, where he directs the company efforts to deliver high quality universal digital pathology software for use in education and biomarker training, life science research and routine diagnostics. His lifelong ambition is to launch products that make a true difference in cancer patient care. We're going to be talking about unified file formats or the lack thereof; How middleware serves to bridge the gap between various components in the digital pathology pixel pathway; What is involved in regulatory certification of software products in digital pathology and how it benefits customers; What steps are needed in achieving various regulatory certifications such as CE IVD or FDA clearance; What are the risks involved in embarking on such an endeavor and what advice Rudy would have for startups or other young companies seeking to develop software products in a regulated environment.

“We shall not cease from exploration and the end of all our exploring will be to arrive where we started and know the place for the first time” - TS Eliot. Is it possible to predict molecular biomarkers from H&E sections through deep learning and computer vision? After all the Hematoxylin and Eosin section was one of our first tools in our arsenal. Molecular diagnostics came center stage in the 1990s and 2000s. Can we know can we now go back to the H&E and predict the molecular features? Our guest is Heather Couture of Pixel Scientia, where she accelerates machine learning projects for pathology startups. She has a PhD in Computer Science and has published in top-tier computer vision and medical imaging venues. She is the host of the Impact AI Podcast and writes regularly on LinkedIn, for her newsletter Pathology ML Insights, and for a variety of other publications. We're going to be talking about artificial intelligence and computer vision in digital pathology. First of all, what do we even mean by artificial intelligence? What is the difference between AI, a computer program, or simply just crunching large amounts of data? How do we develop tools for digital pathology; what are some of the challenges and pitfalls? And yes the “V” word - what do we mean by validation? How do we go about validation in the digital era? And finally, what about those pesky preanalytical factors? Will it even be possible to develop new tools without standardizing or accounting for variability in tissue processing, staining, and scanning?

Amidst this digital transformation in pathology, it’s not only the review of histologic sections that is going digital and moving online, but practically everything else is too. What does this mean for the future of publishing and pathology education with the advent of online content and delivery mechanisms? Our guest is Jason Hornick, MD, PhD, Director of Surgical Pathology and Immunohistochemistry at Brigham and Women's Hospital, Professor of Pathology at Harvard Medical School, a pathologist with expertise in soft tissue tumor pathology, gastrointestinal pathology, and diagnostic immunohistochemistry. He is the Editor-In-Chief of the 5th series of the AFIP Atlases of Tumor and Non-tumor Pathology. Dr. Hornick serves on 14 editorial boards, including the American Journal of Surgical Pathology and Modern Pathology. He is dedicated to the advancement of diagnostic surgical pathology, and teaching approaches to diagnostic pathology to trainees and practitioners We’re going to be talking about the mission and history of the American Registry of Pathology, the publishers of the beloved fascicles atlases of tumor and non-tumor path now in its fifth series. What is the future of publishing going to look like with the advent of online content which can be updated much more frequently than the traditional 5-to-10-year cycle of publishing hard copies of books? What does online content social media and other forces mean for peer review and scientific authority? And what is the future of education going to look like not only for pathology trainees but also education for all pathologists?

One of the biggest pieces of news we've gotten in 2022 is the announcement of new CPT codes from the American Medical Association specifically for digital pathology. What are CPT codes? Why should we care? How will these new codes help us to elevate the practice of anatomic pathology and allow us to more concretely articulate a value proposition for digital pathology? Our guest is Nathan Buchbinder, Chief Product Officer and cofounder at Proscia. We discuss the new CPT codes and much more. We get a current lay of the land for digital pathology in mid-2022. How far have we come in the past five years or so? How far have we come in the past year - which has seen dramatic changes and acceleration in many areas. How do the US and Europe differ in adoption and outlook and what are some continuing barriers to the adoption of digital pathology?

Conquering cancer through Artificial Intelligence is the goal of our guest this week on Digital Pathology today. We're talking with Brandon Suh, MD, CEO of Lunit, a South Korean company creating applications utilizing Artificial Intelligence in both digital pathology as well as radiology. Lunit stands for “learning unit,” which is fundamental to the company's philosophy which embodies the belief that they can learn just about anything and is foundational to the Artificial intelligence, which involves continuous learning. The company has developed algorithms focusing on disease screening in the areas of mammography and chest X-ray in radiology and predictive and prognostic markers in digital pathology - in particular an AI based algorithm for immunotherapy, which outperforms the current standard of care. Lunit is looking to capitalize on the growing trend of Artificial Intelligence in medicine.

As we kick off this third season of digital pathology today our guest is Dr. Kevin Sandeman, from the Region Skane in Sweden, who is going to be taking us through their digital transformation. The Region Skane covers a population of over 1.4 million people, involves four health care systems including two academic and two community centers. The caseload is over 110,000 cases per year - generating over 680,000 scanned slides. This is a wonderful story, and the benefits are maybe not what you would expect. In addition to the obvious benefits of not having to he tethered to a microscope, according to Dr. Sandeman, the number one benefit resulting from the digital transformation is “Peace of Mind.” The departments in the region enjoyed much higher quality in terms of histologic preparations and output, enhanced engagement and understanding among the histotechnologists, reduction of hystertechnologist FTE requirements and complete elimination of the need to work overtime. Turnaround time for case reporting has dropped an astounding 33% and clinician satisfaction is remarkably increased. We also talked about some of the challenges in adopting the digital workflow not to mention the IT infrastructure requirements and storage demands but at the top of the list appeared to be the initial pathologist resistance to adopting the digital workflow.

The digital transformation in pathology is well underway but what kind of opportunities does this present to us? We've heard all the buzz words such as machine learning artificial intelligence and heard all the hype about enhanced workflows and making pathologists’ lives easier. What is the future practice of pathology going to look like? Our guest is Beatrice Knudsen, MD, PhD, Medical Director of Digital and Computational Pathology at ARUP Laboratories, as well as Professor of Pathology at the University of Utah. Dr. Knudsen has built a program that integrates histopathology, multiplex tissue staining, digital image analysis, as well as computational pathology. She is applying this approach to close the gap between laboratory research and clinical care and to develop novel algorithms for diagnosis, prognosis and the treatment of patients.

They're coming - AI and machine learning algorithms in digital pathology. How do we go about assessing the robustness and suitability of these new tool for use in actual clinical practice? Our guests are Dr. Angelique Levi, Associate Professor of Pathology and Vice Chair and Director of Pathology Outreach Services and Dr. Sudhir Perincheri, Assistant Professor of Pathology and Director of Digital Pathology, both from Yale School of Medicine. We're talking about how new tools are going to be incorporated into clinical practice, and the impact of these new tools in various practice settings, such as academic medical centers, community outreach, and private practice. What does this mean for the training of the next generation of pathologists? We're only human, after all, but what about machine algorithms? What level of mistakes or "misses" is allowable? And what if these tools are too sensitive, could they actually slow us down and make life more complicated for us?

Too many systems or components in a pathology workflow adds complexity and cost. How can we create an elegant end to end solution to enhance workflows, while decreasing both complexity and costs? Our guest is John Wirthlin, CEO of Lumea, a company reimagining pathology - beginning and ending with the patient. We discuss the end-to-end solution in digital pathology. One of the promises of digital pathology is making the pathologist’s workflow more streamlined and more efficient. But can we bring the Laboratory Information System (LIS) into the clinic and improve the workflow flow for our clinician colleagues as well. How can we improve the tissue handling or preanalytic phase of testing? It's often easy to forget who the customer is. By utilizing enhanced workflows, can we improve satisfaction for the clinician, who is ultimately treating the patient?

Amidst this digital transformation in pathology the talk seems to be largely centered around whole slide imaging of H&E-stained tissue sections. But what about cytology? Why don't we hear much talk about digital cytology, and what are some of the unique challenges in cytology in terms of going digital. Our guest is Michael Quick, Vice President of Research and Development and Innovation at Hologic. We discuss the history and evolution of cervical cancer screening all the way from pap smears to thin prep to incorporating digital pathology and image analysis in the process. Will we finally be able to say goodbye to the light microscope? And what are some of the parallels and lessons learned comparing cervical cancer screening by cytology in pathology to breast cancer screening by digital mammography in radiology.

What is going to surprise us in digital pathology in big ways and in small ways? How can seemingly small changes make a big impact in what we do as pathologists and in the lives of patients? How are we going to make use of the large datasets that we will inevitably be generating through the widescale adoption of digital pathology? Our guest is Louis Culot, General Manager for Genomics and Oncology Informatics at Phillips. We're going to be talking about the current landscape of digital pathology in 2022, where we've been in the past 20 years and even earlier, and the impact of the COVID-19 global health emergency on the acceleration of the adoption of digital pathology. And what is the future of integrated diagnostics, incorporating digital pathology radiology and molecular diagnostics.

It's easy to get caught up in the hype surrounding digital pathology. But what is actually involved and going digital? What are the steps involved? We know about scanners and viewing images, of course. But what other pieces of software might be necessary or helpful? Coleman Stavish is co-founder and CTO at Proscia, a leading provider of digital and computational pathology solutions. At Proscia, Coleman works with a team of engineers and scientists who develop enterprise software in use at leading laboratories and large biopharma companies around the world, as well as novel image classification. We're going to be talking about workflows in digital pathology and the importance of the image management platform. What are the considerations in going digital, how can all of our software work together and how can we truly realize the promise of digital pathology by incorporating AI algorithms and other tools?

NVIDIA is involved in computer graphics, artificial intelligence, computing and processors across variety of industries including gaming, entertainment, finance, and health care. What are the commonalties of all these industries and what lesson from other fields can we apply to healthcare? Our guest is Craig Rhodes EMEA Industry Lead - AI for Healthcare and Life Sciences at NVIDIA. There is the perception that healthcare lags other industries in terms of AI applications. Is this true or is it a misconception? How is AI changing healthcare now and for the future How can we accelerate the adoption of AI and new technologies in health care? NVIDIA has launched Cambridge-1 - a powerful supercomputer in the UK. We will learn about this and what results we can expect from Cambridge one in the difference this will make in healthcare in the lives of patients.

Our guests are Dr. Garth Bryson Consultant Pathologist and Clinical Director for Laboratory Medicine at the NHS Greater Glasgow and Clyde and Dr. David Harrison, Professor of Pathology from the University of Saint Andrews, and both from ICAIRD - the Industrial Center for Artificial Intelligence Research and Digital Diagnostics a program matching AI solutions and research with health and social care priorities across Scotland. We talk about the long journey of implementing digital pathology in a large health care system, barriers to adoption of digital pathology, the promise of artificial intelligence, the business case for digital pathology – have we finally been able to articulate one? And finally, we discuss our guests’ experience so far in creating a digital slide sharing network throughout Scotland.