Digital Pathology Podcast

Aleksandra Zuraw, DVM, PhD
  • 0.0 (0)
  • NATURAL SCIENCES
  • UPDATED WEEKLY

Aleksandra Zuraw from Digital Pathology Place discusses digital pathology from the basic concepts to the newest developments, including image analysis and artificial intelligence. She reviews scientific literature and together with her guests discusses the current industry and research digital pathology trends.

  1. 6 DAYS AGO

    196: DigiPath Digest #39 - If AI Sees More Than We Do. What Makes It Clinically Trustworthy?

    Send a text If AI can detect patterns we cannot see, how do we know when its answers are clinically trustworthy? In this episode of DigiPath Digest #39, I explore a big-picture question in digital pathology and medical AI. Many models now match or even exceed human performance in specific diagnostic tasks. But most of that evidence comes from controlled or retrospective datasets. So what happens when we try to bring these tools into real clinical workflows? I review four recent papers that help frame this challenge and point toward the next steps for trustworthy AI in healthcare.  You will hear about the role of prospective validation, real-world effectiveness, transparent reporting standards, and multimodal data integration as recurring themes across these studies. Key Highlights 00:00 – Introduction What do we do when AI detects signals that humans cannot see? The core challenge is verifying those outputs before trusting them in clinical decision making.  03:32 – AI Across the Healthcare Continuum A narrative review shows AI achieving clinician-level performance in well-defined imaging tasks, including digital pathology. But most evidence comes from retrospective or controlled environments, and prospective validation remains limited.  08:34 – Multi-Omics and AI in Gastric Biopsy Diagnostics Morphology alone cannot fully capture molecular heterogeneity or predict disease progression. Integrating genomics, proteomics, metabolomics, and other omics with AI is shifting gastric pathology toward data-driven precision gastroenterology.  13:38 – Hyperspectral Imaging for Real-Time Surgical Guidance Spectral imaging can analyze tissue composition during surgery without staining, freezing, or contact with the tissue. Studies show promising sensitivity for detecting malignancy and supporting intraoperative decision making.  17:20 – REFINE Reporting Guideline for Foundation Models and LLMs An international consensus guideline introduces a 44-item reporting checklist to standardize how AI studies are described. The goal is transparent, reproducible, and comparable research in medical AI.  22:35 – Big Takeaway AI should be viewed as clinical decision support, not a replacement for clinicians. Real-world validation, ethical governance, and reproducible research standards will determine how these tools enter pathology workflows.  References (Articles Discussed) Artificial Intelligence in Healthcare: From Diagnosis to Rehabilitation  https://pubmed.ncbi.nlm.nih.gov/41755929/ Transforming Gastric Biopsy Diagnostics: Integrating Omics Technologies and Artificial Intelligence  https://pubmed.ncbi.nlm.nih.gov/41751306/ From Image-Guided Surgery to Computer-Assisted Real-Time Diagnosis with Hyperspectral and Multispectral Imaging  https://pubmed.ncbi.nlm.nih.gov/41750768/ REFINE Reporting Guideline for Foundation and Large Language Models in Medical Research  https://pubmed.ncbi.nlm.nih.gov/41762555/ If you enjoy staying current with digital pathology and AI research, this episode will help you connect the dots between promising algorithms and practical clinical adoption. Support the show Get the "Digital Pathology 101" FREE E-book and join us!

    27 min

  2. 2 MAR

    191: Hallucinations, Agents, and AI in Pathology

    Send a text Clinical Artificial Intelligence in 2026. Accuracy, Education, and Guardrails Artificial intelligence is evolving fast in medicine. But how accurate is it. And are we building it safely? In this episode of DigiPath Digest, I review five new studies shaping digital pathology, radiology, burn diagnostics, and agent-based large language model systems. We discuss accuracy gains, hallucination filtering, education challenges, and why safeguards are essential before clinical deployment. Clear. Practical. Evidence-based. ⏱ Topics & Timestamps [00:02] Introduction Weekly journal club on digital pathology and artificial intelligence. [05:13] Hallucination Filtering in Radiology Using Discrete Semantic Entropy to detect hallucination-prone responses in Vision Language Models. Accuracy improved from 51.7 percent to 76.3 percent after filtering high-entropy answers. [15:04] Artificial Intelligence in Pathology Training Supervised use during residency. Balancing artificial intelligence adoption with preservation of morphological analysis and critical thinking. [20:12] Colorectal Cancer Lymph Node Detection Two-stage classification and segmentation model in Whole Slide Imaging. Recall 1.0. Specificity 0.935. Dice coefficient 0.818. Artificial intelligence as a second opinion. [25:04] Burn Depth Prediction with Artificial Intelligence Tissue Doppler Elastography and Harmonic B-mode ultrasound combined with artificial intelligence. 90 to 95 percent accuracy in human subjects. [31:20] Agent-Based Large Language Model Systems OpenManus and Manus evaluated in clinical simulations. Up to 60.3 percent accuracy. High computational cost. 89.9 percent of hallucinations filtered by safeguards. [40:08] Patient Access to Pathology Images Why viewing pathology slides can empower patients and improve communication. Resources https://pubmed.ncbi.nlm.nih.gov/41720937/https://pubmed.ncbi.nlm.nih.gov/41720644/https://pubmed.ncbi.nlm.nih.gov/41716065/https://pubmed.ncbi.nlm.nih.gov/41709317/https://pubmed.ncbi.nlm.nih.gov/41708802/Support the show Get the "Digital Pathology 101" FREE E-book and join us!

    30 min

  3. 24 FEB

    190: Can a Better Stain Improve AI in Pathology?

    Send a text What if one of the biggest sources of diagnostic variability in prostate cancer isn’t the pathologist—but the stain we’ve trusted for decades? In this episode, I speak with Professor Ingid Carlbom, founder of CADESS.AI, about a different way to approach prostate cancer grading—by rethinking staining, segmentation, and AI decision support from the ground up. We explore why 30–40% interobserver variability persists in Gleason grading and how optimized stains combined with explainable AI can significantly reduce that uncertainty. Ingrid shares her journey from applied mathematics and computer science into pathology, the skepticism she faced in 2008, and why CADESS.AI chose not to “optimize H&E,” but instead developed a Picrosirius red + hematoxylin stain designed specifically for computational pathology. We discuss how grading at the gland and cellular level improves reproducibility, why explainability matters for trust, and what it really takes to build both stain and software as a single diagnostic workflow. This conversation challenges long-held assumptions—and asks whether improving data quality should come before building smarter algorithms. Highlights: [00:00–01:08] The problem: 30–40% disagreement in prostate cancer grading[01:08–03:03] Ingrid’s path from applied math to digital pathology[03:03–04:58] Early skepticism toward AI in pathology and fear of replacement[04:58–08:56] Why H&E limits segmentation—and how a new stain changes that[10:55–15:09] Clinical testing: non-inferiority, AI assistance, and NCCN risk stratification[19:47–22:59] Explainable UI: color-coded glands and pathologist override[26:16–27:29] Why grading glands (not whole slides) reduces variability[38:09–41:47] Regulatory challenges of combined stain + AI devices[45:52–48:55] The future of optimized stains in routine pathology Resources from This Episode CADESS.AI – Prostate cancer decision support systemNCCN prostate cancer risk stratification guidelinesSupport the show Get the "Digital Pathology 101" FREE E-book and join us!

    56 min

  4. 24 FEB

    189: Digital Pathology Deployment Decoded the Rigorous 4 Phase Framework

    Send a text Sometimes a paper comes out that’s so practical and relevant to what we do in digital pathology that I know we have to talk about it. In this episode, I dive into “A Guide for the Deployment, Validation and Accreditation of Clinical Digital Pathology Tools” from Geneva University Hospital (HUG) — one of the most useful, real-world frameworks I’ve seen for bringing digital pathology tools safely into clinical practice. If you’ve ever built an AI model and wondered, “Now what?”, this episode is for you. Because building the model is often the easy part — deployment is where things get complex. This guide breaks the process into four practical phases every lab can follow: 1️⃣ Pre-Development – Define your clinical need, project scope, and validation plan before writing a single line of code. 2️⃣ Development – Build and integrate the algorithm in a production-ready environment. 3️⃣ Validation & Hardening – Turn your research code into a reliable, secure, and compliant clinical tool. 4️⃣ Production & Monitoring – Keep the tool validated and performing consistently over time. We also discuss what makes qualification, validation, and accreditation different — and why that order really matters. You’ll hear about the multidisciplinary team behind these deployments, especially the deployment engineer (DE) — the technical linchpin who turns AI research into clinical reality. I share the story of HUG’s H. pylori detection tool, which cut diagnostic time by 26% while maintaining a 0% false negative rate. The team’s secret? Careful planning, quality control, and continuous user feedback — not just great code. Other highlights include: Why integration often takes longer than building the AI model itselfHow to avoid invalidating your validation dataWhat continuous performance monitoring looks like in real labsAnd why every lab still needs to do local validation, even with proven toolsIf you’re working on digital or computational pathology tools — or just want to understand how AI safely moves from research to routine diagnostics — this episode will give you a roadmap grounded in real experience. 🎧 Listen now to learn how to move from algorithm to accreditation, step by step. And if you’re just getting started in digital pathology, I’d love to give you my free eBook, Digital Pathology One-on-One: All You Need to Know to Start and Continue Your Digital Pathology Journey. You’ll find the link to download it in the show notes. See you in the episode! Support the show Get the "Digital Pathology 101" FREE E-book and join us!

    23 min

  5. 21 FEB

    188: AI in Pathology: Biomarkers, Multimodal Data & the Patient

    Send a text Is AI in pathology actually improving diagnosis — or just adding complexity? In DigiPath Digest #37, we reviewed four recent publications covering AI-based biomarker quantification in glioblastoma, real-world digital workflow integration in prostate cancer, multimodal AI combining histopathology and genomics, and patient perspectives on AI in cancer diagnostics. This episode connects technical performance with something equally important: trust. Episode Highlights [00:02] Community & updates Digital Pathology 101 free PDF, upcoming patient-focused book, and global attendance. [04:07] AI-based image analysis in glioblastoma AI showed strong consistency with pathologists when quantifying Ki-67, P53, and PHH3. Significant biological correlations (Ki-67 ↔ PHH3, PHH3 ↔ P53) were detected by AI — not by manual assessment. Takeaway: computational quantification improves precision. [09:28] Real-world digital workflow + AI in prostate cancer (France) AI-pathologist concordance: • 93.2% (high probability cancer detection) • 99.0% (low probability slides) Gleason concordance: 76.6% 10% failure rate due to pre-analytical artifacts. Takeaway: infrastructure and sample quality still matter. [15:58] Multimodal AI (MARBIX framework) Combines whole slide images + immunogenomic data in a shared latent space using binary “monograms.” Performance in lung cancer: 85–89% vs 69–76% unimodal models. Takeaway: integrated data improves case retrieval and similarity reasoning. [22:13] AI-powered paper summary subscription introduced Structured summaries for busy professionals who want more than abstracts. [26:17] Patient roundtable on AI in pathology (Belgium) Patients expect: • Better accuracy • Faster turnaround • Stronger collaboration Trust is high when:  • Algorithms use diverse datasets  • Pathologists retain final responsibility Clinical validity mattered more than full algorithm transparency.  Privacy concerns focused more on insurer misuse than cloud transfer. Key Takeaways AI improves biomarker precision in glioblastoma.Digital pathology implementation works — but pre-analytics can limit AI performance.Multimodal AI represents the next meaningful step in precision diagnostics.Patients are not afraid of AI — they want validation, oversight, and governance.Human–AI collaboration remains central.If you’re working in digital pathology, computational pathology, or precision oncology, this episode connects evidence, implementation, and patient perspective. Support the show Get the "Digital Pathology 101" FREE E-book and join us!

    21 min

  6. 20 FEB

    184: Digital Pathology Guidelines: What Every Lab Must Get Right

    Send a text What actually needs to be in place before digital pathology can replace the microscope? In this episode of DigiPath Digest, I walk through the 2026 Polish Society of Pathologists guidelines and translate them into practical steps for real pathology labs. This isn’t theory. It’s about hardware fidelity, data integrity, validation, and AI integration — and what each of these actually requires in daily workflow. We talk about scanner resolution standards (≤0.26 μm per pixel), 4K monitor calibration, visually lossless compression (20:1), scalable storage, pathologist-driven validation, and what “non-inferiority” truly means. Digital pathology is not just a change of medium. It’s an operational shift. Episode Highlights [00:02] Community & growth 1,600+ new newsletter subscribers, 10,000+ Facebook members, and free Digital Pathology 101 book access. [07:20] The 4 pillars of adoption Hardware fidelity · Data integrity · Clinical validation · Future integration. [08:30] Hardware requirements 40x equivalent scanning (≤0.26 μm/px), 4K monitors, >300 cd/m² luminance, 10-bit color depth. [12:00] Workflow & throughput 200–300 slides/day per scanner, automated focus control, urgent case prioritization. [17:25] Storage & archiving ~1 GB per slide. Active archive (6–24 months). Long-term retention (10–20 years). GDPR compliance & TLS encryption. [23:09] Validation philosophy Pathologist-centered validation. Two phases: • Familiarization (~20 retrospective cases) • Dual review with discrepancy tracking Goal: digital must be non-inferior to glass. [29:03] AI in digital pathology AI supports quantification (Ki-67, HER2, ER/PR, PD-L1), tumor detection, and future multimodal predictions — but pathologists remain central. [33:26] Intraoperative telepathology 5-minute scan-to-view time. Minimum 100 Mbps upload. Redundancy and safety protocols required. [34:50] Can digital cameras replace scanners? Hybrid workflows exist. Regulatory compliance still applies. [38:19] Adoption checklist summary Certified scanners (CE-IVD/FDA), calibrated monitors, scalable storage, phased validation, and documented QC. Key Takeaways Digital pathology adoption is a structured process — not just buying a scanner.Validation is individualized and tissue-specific.Infrastructure and quality control are as important as image quality.AI enhances reproducibility and quantification but does not replace pathologists.Regulatory compliance and data governance are non-negotiable.Support the show Get the "Digital Pathology 101" FREE E-book and join us!

    34 min

  7. 8 FEB

    182: AI, Quality, and Standards: The Next Chapter of Digital Pathology

    Send a text This session is a practical walkthrough of where digital pathology and AI truly stand in early 2026—based on five recent PubMed papers and real-world implementation experience. In this episode, I review new clinical adoption guidelines, AI applications in liver cancer imaging and pathology, AI-ready metadata for whole slide images, non-destructive tissue quality control from H&E slides, and machine learning–assisted IHC scoring in precision oncology. This conversation is not about hype. It’s about standards, validation, data integrity, and clinical translation—the factors that decide whether AI tools stay in research or reach patient care. Episode Highlights 01:21 – Practical digital pathology adoption guidelines (Polish Society of Pathologists)08:05 – AI in liver cancer imaging & pathology, and why framework alignment matters18:10 – AI-generated tissue maps as metadata for WSI archives23:01 – PathQC: predicting RNA integrity and autolysis from H&E slides32:14 – ML-assisted IHC scoring in genitourinary cancers29:42 – Digital Pathology 101 book + community updatesKey Takeaways Digital pathology adoption still requires clear standards and validation workflowsAI performs best when aligned with existing diagnostic frameworks (e.g., LI-RADS)Metadata extraction is a low-effort, high-impact AI use caseSlide-based quality control can support biobanking and biomarker researchAutomated IHC scoring improves consistency—but adoption remains uneven globallyResources Mentioned  Digital Pathology 101 (free PDF & audiobook)Publication Links:  a. https://pubmed.ncbi.nlm.nih.gov/41618426/                                                                 b. https://pubmed.ncbi.nlm.nih.gov/41616271/                                                                   c. https://pubmed.ncbi.nlm.nih.gov/41610818/                                                                 d. https://pubmed.ncbi.nlm.nih.gov/41595938/                                                                 e. https://pubmed.ncbi.nlm.nih.gov/41590351/  Support the show Get the "Digital Pathology 101" FREE E-book and join us!

    26 min

  8. 24 JAN

    181: Can AI Read Clinical Text, Tissue, and Costs Better Than We Can?

    Send a text What happens when artificial intelligence moves beyond images and begins interpreting clinical notes, kidney biopsies, multimodal cancer data, and even healthcare costs? In this episode, I open the year by exploring four recent studies that show how AI is expanding across the full spectrum of medical data. From Large Language Models (LLM) reading unstructured clinical text to computational pathology supporting rare kidney disease diagnosis, multimodal cancer prediction, and cost-effectiveness modeling in oncology, this session connects innovation with real-world clinical impact. Across all discussions, one theme is clear: progress depends not just on performance, but on integration, validation, interpretability, and trust. HIGHLIGHTS: 00:00–05:30 | Welcome & 2026 Outlook New year reflections, global community check-in, and upcoming Digital Pathology Place initiatives. 05:30–16:00 | LLMs for Clinical Phenotyping How GPT-4 and NLP automate phenotyping from free-text EHR notes in Crohn’s disease, reducing manual chart review while matching expert performance. 16:00–23:30 | AI Screening for Fabry Nephropathy A computational pathology pipeline identifies foamy podocytes on renal biopsies and introduces a quantitative Zebra score to support nephropathologists. 23:30–29:30 | Is AI Cost-Effective in Oncology? A Markov model evaluates AI-based response prediction in locally advanced rectal cancer, highlighting when AI delivers value—and when it does not. 29:30–38:30 | LLM-Guided Arbitration in Multimodal AI A multi-expert deep learning framework uses large language models to resolve disagreement between AI models, improving transparency and robustness. 38:30–44:30 | Real-World AI & Cautionary Notes Ambient clinical scribing in practice, AI hallucinated citations, and why guardrails remain essential. KEY TAKEAWAYS • LLMs can extract meaningful clinical phenotypes from narrative notes at scale  • AI can support rare disease diagnosis without replacing expert judgment  • Economic value matters as much as technical performance  • Explainability and arbitration are becoming critical in multimodal AI systems  • Human oversight remains central to responsible adoption Resources & References Digital Pathology Place: https://www.digitalpathologyplace.comDigital Pathology 101 (free PDF, updates included)Automating clinical phenotyping using natural language processingZebra bodies recognition by artificial intelligence (ZEBRA): a computational tool for Fabry nephropathyCost-effectiveness analysis of artificial intelligence (AI) for response prediction of neoadjuvant radio(chemo)therapy in locally advanced rectal cancer (LARC) in the NetherlandsA multi-expert deep learning framework with LLM-guided arbitration for multimodal histopathology predictionSupport the show Get the "Digital Pathology 101" FREE E-book and join us!

    35 min
See All (188)

Trailer

About

Aleksandra Zuraw from Digital Pathology Place discusses digital pathology from the basic concepts to the newest developments, including image analysis and artificial intelligence. She reviews scientific literature and together with her guests discusses the current industry and research digital pathology trends.

Information

  • Creator
    Aleksandra Zuraw, DVM, PhD
  • Years Active
    2019 - 2026
  • Episodes
    188
  • Rating
    Clean
  • Copyright
    © 2026 Digital Pathology Podcast
  • Show Website
    Digital Pathology Podcast

You Might Also Like