PathAI

✈️ PathAI is heading to NeurIPS and Machine Learning for Health (ML4H), both in San Diego, CA Our team will be sharing new research at the intersection of machine learning and pathology – and we’d love to connect while we’re there. 𝐍𝐞𝐮𝐫𝐈𝐏𝐒 𝐈𝐦𝐚𝐠𝐞𝐨𝐦𝐢𝐜𝐬 𝐖𝐨𝐫𝐤𝐬𝐡𝐨𝐩 – 𝐈𝐦𝐚𝐠𝐞𝐨𝐦𝐢𝐜𝐬: 𝐃𝐢𝐬𝐜𝐨𝐯𝐞𝐫𝐢𝐧𝐠 𝐁𝐢𝐨𝐥𝐨𝐠𝐢𝐜𝐚𝐥 𝐊𝐧𝐨𝐰𝐥𝐞𝐝𝐠𝐞 𝐟𝐫𝐨𝐦 𝐈𝐦𝐚𝐠𝐞𝐬 𝐔𝐬𝐢𝐧𝐠 𝐀𝐈 1️⃣Poster: 𝘌𝘷𝘢𝘭𝘶𝘢𝘵𝘪𝘯𝘨 𝘵𝘩𝘦 𝘜𝘵𝘪𝘭𝘪𝘵𝘺 𝘰𝘧 𝘚𝘱𝘢𝘳𝘴𝘦 𝘈𝘶𝘵𝘰𝘦𝘯𝘤𝘰𝘥𝘦𝘳𝘴 𝘧𝘰𝘳 𝘐𝘯𝘵𝘦𝘳𝘱𝘳𝘦𝘵𝘪𝘯𝘨 𝘢 𝘗𝘢𝘵𝘩𝘰𝘭𝘰𝘨𝘺 𝘍𝘰𝘶𝘯𝘥𝘢𝘵𝘪𝘰𝘯 𝘔𝘰𝘥𝘦𝘭 Presenter: Syed Ashar Javed Poster presentation- Dec 6 |  8:45–9:30 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐟𝐨𝐫 𝐇𝐞𝐚𝐥𝐭𝐡𝐜𝐚𝐫𝐞 𝐒𝐲𝐦𝐩𝐨𝐬𝐢𝐮𝐦 (𝐌𝐋4𝐇)  1️⃣ Spotlight Talk and Poster: 𝘊𝘢𝘯𝘤𝘦𝘳 𝘪𝘥𝘦𝘯𝘵𝘪𝘧𝘪𝘤𝘢𝘵𝘪𝘰𝘯 𝘢𝘯𝘥 𝘴𝘪𝘮𝘶𝘭𝘢𝘵𝘦𝘥 𝘤𝘢𝘴𝘦 𝘱𝘳𝘪𝘰𝘳𝘪𝘵𝘪𝘻𝘢𝘵𝘪𝘰𝘯 𝘸𝘪𝘵𝘩 𝘛𝘶𝘮𝘰𝘳𝘋𝘦𝘵𝘦𝘤𝘵™ Presenter: Nishant Agrawal Oral presentation- Dec 2 | 10:05–10:25 Poster presentation- Dec 2 | 15:25–16:25 2️⃣ Poster: 𝘊𝘰𝘯𝘧𝘰𝘳𝘮𝘢𝘭 𝘱𝘳𝘦𝘥𝘪𝘤𝘵𝘪𝘰𝘯𝘴 𝘧𝘰𝘳 𝘮𝘶𝘭𝘵𝘪-𝘤𝘭𝘢𝘴𝘴 𝘥𝘦𝘳𝘮𝘢𝘵𝘰𝘱𝘢𝘵𝘩𝘰𝘭𝘰𝘨𝘺 𝘸𝘪𝘵𝘩 𝘗𝘢𝘵𝘩𝘈𝘴𝘴𝘪𝘴𝘵 𝘋𝘦𝘳𝘮™ Presenter: Nishant Agrawal Poster presentation- Dec 1 | 15:30–16:30 🚀 We’re hiring! Passionate about AI’s impact on healthcare? We're hiring at PathAI! Join Harsha Vardhan Pokkalla, Nishant Agrawal, and Syed Ashar Javed at NeurIPS and ML4H to talk about our open roles and innovations in pathology, diagnostics, and beyond. Don't miss the chance to connect! Reach out directly to set up a time to chat during the conferences. #NeurIPS #ML4H #MachineLearning #AI #DigitalPathology #Hiring #PathAI *𝘛𝘶𝘮𝘰𝘳𝘋𝘦𝘵𝘦𝘤𝘵™ 𝘢𝘯𝘥 𝘗𝘢𝘵𝘩𝘈𝘴𝘴𝘪𝘴𝘵 𝘋𝘦𝘳𝘮™ 𝘢𝘳𝘦 𝘍𝘰𝘳 𝘙𝘦𝘴𝘦𝘢𝘳𝘤𝘩 𝘜𝘴𝘦 𝘖𝘯𝘭𝘺. 𝘕𝘰𝘵 𝘧𝘰𝘳 𝘶𝘴𝘦 𝘪𝘯 𝘥𝘪𝘢𝘨𝘯𝘰𝘴𝘵𝘪𝘤 𝘱𝘳𝘰𝘤𝘦𝘥𝘶𝘳𝘦𝘴.

🔬Assessing fibrosis in MASH clinical trials is limited by semi-quantitative scoring criteria. What if you could better understand therapeutic-induced changes in liver fibrosis composition beyond the current categorical staging? 💡This is the transformative potential of LiverExplore,* our suite of deep-learning models trained to exhaustively predict and quantify features of the liver microarchitecture – including fibrosis – directly from digitized H&E-stained liver biopsies. 🚀 At the recent AASLD conference, PathAI collaborators from Akero Therapeutics and Altimmune, Inc. presented results from two clinical trials showing how LiverExplore can dissect changes in fibrosis that result from treatment. 📃 Read about these data and how LiverExplore can improve our understanding of MASH clinical trial results in our latest blog. *LiverExplore is For Research Use Only. Not for use in diagnostic procedures. #PathAI #DigitalPathology #AI #ML #ComputerVision #ComputationalPathology #AASLD #Liver #Fibrosis

🧠 A foundation model’s intelligence lives in its embeddings, how it “understands” and encodes complex visual patterns. 📖 In our latest blog, 𝐁𝐞𝐲𝐨𝐧𝐝 𝐭𝐡𝐞 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧: 𝐔𝐬𝐢𝐧𝐠 𝐏𝐋𝐔𝐓𝐎 𝐄𝐦𝐛𝐞𝐝𝐝𝐢𝐧𝐠𝐬 𝐟𝐨𝐫 𝐌𝐨𝐝𝐞𝐥 𝐅𝐢𝐧𝐞-𝐭𝐮𝐧𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝐒𝐢𝐦𝐢𝐥𝐚𝐫𝐢𝐭𝐲 𝐒𝐞𝐚𝐫𝐜𝐡, we dive deeper into PLUTO’s embeddings and show how they power a similarity search within our platform to reveal image tiles with similar histologic features. This capability allows us to:  - Mine data with rare model failure modes,  - Target and supercharge training data curation  - Unlock smarter model fine-tuning that boosts performance on challenging histologies. From tile-level representations of whole-slide images to scalable model training workflows for pathologists and ML teams, this post explores how embeddings move from backbone to game-changer. 🔗 Read more here >>

🎉 We’re excited to see new real-world data from the Universitätsspital Zürich (USZ) highlighting the impact of AI-based tumor cell quantification in routine molecular diagnostics. At this week’s 90th Annual Congress of the Swiss Society of Pathology, the USZ team will present results using PathAI’s AIM-TumorCellularity (AIM-TC). AIM-TC is PathAI's workflow algorithm for tumor cell content estimation, available within the CE-IVDR-certified digital pathology platform, AISight Dx. This study represents the most comprehensive real-world validation of a tumor cell content algorithm to date. If you’re attending the 90th Annual Congress of the Swiss Society of Pathology this week in Zürich, make sure to visit Poster A10 and join the presentation by Mariam Gachechiladze MD, PhD, to learn more. 📌 Poster A10: “Impact of preanalytical and analytical factors on AI-based tumor cell content quantification in digitalized diagnostic samples for molecular tumor profiling: A Real-World Study” 🗓️ Presentation times: Thursday, Nov 20, 14:45-15:15 and Friday, Nov 21, 09:30-10:30 The study reports several encouraging findings: - Stronger agreement with NGS-based (FMI) tumor cell estimates than pathologist-derived assessments, across multiple tumor types. - High analytical performance of AIM-TC with Sensitivity 0.98, Specificity 0.94, and Accuracy 0.97 at the single-cell level when reviewed by pathologists. - Robust performance across biopsies, resections, and cytology, with clear insights into how pre- and analytical factors influence variability. We are proud to see AIM-TC and AISight Dx supporting high-quality, standardized tumor cell assessment in a real-world setting, and grateful to the USZ team for advancing the understanding of AI in molecular pathology. Mariam Gachechiladze MD, PhD, Jan-Niklas Runge, Jan H. Rüschoff, Viktor H. Koelzer, Bettina Sobottka *AISight Dx is CE-IVD-marked for primary diagnosis in the EEA, CH, and UK. AIM-TumorCellularity is an AISight Dx workflow tool and not intended to be used for diagnostic purposes in the EEA/CH/UK and is Research Use Only in the US.

📣 PathAI announces self-service integration of PathExplore™ on AISight® IMS to advance tumor microenvironment analysis, with early access available to select academic medical centers. 🔬 This early access release empowers academic medical centers to explore the tumor–immune landscape at scale, accelerating discovery in immuno-oncology and pathology research while leveraging PathAI’s AI-powered digital pathology platform. Inside AISight, teams can extract quantitative, human-interpretable features from routine H&E whole-slide images, visualize tumor segmentation and immune cell predictions, quantify tumor-infiltrating lymphocyte (TIL) density, and run cohort-scale analyses in the browser. The cloud-native system integrates with major laboratory information systems for rapid activation without new on-premises infrastructure and helps multi-site studies stay aligned and reproducible. Read the announcement: https://lnkd.in/d2iThNtc 📊 Learn more about PathExplore or AISight: contact digital.dx@pathai.com or visit https://lnkd.in/dTANx9F5 PathExplore is powered by PLUTO, PathAI’s leading foundation model purpose-built for the next generation of AI pathology solutions. Learn more about PLUTO: https://lnkd.in/dWRiWbBR *PathExplore™ and AISight® are for Research Use Only. Not for use in diagnostic procedures. #PathExplore #AISight #AIPathology #TumorMicroenvironment #TME #HandE #TranslationalResearch #Biomarkers #ImmunoOncology #AcademicMedicalCenters

📣 PathAI is heading to Carrefour Pathologie 2025 in Paris! Come to meet us at Booth 13 from November 19 to November 21, 2025 at CNIT Forest, Paris–La Défense, to explore how AI is shaping the future of digital pathology. We look forward to showcasing how the AISight® Image Management System and our AI integrations are advancing digital pathology, and the role of AI and machine learning in digital diagnostics. ✨On se voit à Paris! ✨ 🤝 To schedule a meeting at Carrefour with PathAI: digitaldx@pathai.com 🔗 Event details: https://lnkd.in/df7W6BB  #DigitalPathology #AISight#ImageManagementSystem #PathAI *AISight® is for Research Use Only in the US, AISight® Dx is CE-IVD marked in the EEA, UK, and Switzerland.

This year, The Boston Globe’s 5th Globe Summit brings together revolutionary ideas and leaders shaping the future. 🌎✨ Our CEO Andrew Beck will join the Healthtech and AI panel to discuss how artificial intelligence is transforming healthcare innovation and patient outcomes, an area that lies at the heart of PathAI’s mission to improve patient care through advanced digital pathology. 📅 Date/Time: Nov 18, 2025 · 11:40 AM–12:00 PM ET 📍 Location: The Boston Globe Global Summit 2025, Boston 🔗 Learn more: https://lnkd.in/eKVxYBzy #Healthtech #PathAI #DigitalPathology #Innovation #Boston #GlobeSummit2025 #BostonGlobeMedia

Congratulations to the team, we are thrilled to see PathAI data contributing to this important work. 🎉 🔎 Dive into the power of fibrosis quantification with LiverExplore in our new blog: https://lnkd.in/ex-HtaCY LiverExplore is For Research Use Only. Not for use in diagnostic procedures. #PathAI #LiverExplore #Fibrosis #DigitalPathology #AIinHealthcare #LiverDisease

✈️ Wheels down in Washington, DC for #TheLiverMeeting! We’re ready to share PathAI’s AI-driven liver portfolio—visit 📍 Booth 926 to discover more about the newly EMA-qualified AIM-MASH AI Assist, experience a live LiverExplore™ demo, and learn more about how we are reshaping MASH clinical trials and digital pathology. 📖 Dive into our latest blog on LiverExplore™ and how it’s powering more objective, scalable histology insights: Read the blog here: https://lnkd.in/gMTwGMSf) 🎙️Catch up on our research presented with collaborators during the conference here: https://lnkd.in/gnhqwZm5 🤝 See you at Booth 926—let’s talk biomarkers, endpoints, and real-world trial impact. *LiverExplore and AIM-MASH are For Research Use Only. Not for use in diagnostic procedures. #AIinHepatology #DigitalPathology #MASH #MASLD #LiverDisease #Biomarkers #ClinicalTrials

🔎 Liver fibrosis assessment is complicated by interpretation variability and semi-quantitative measurements. What if you could gain comprehensive, quantitative insights into liver fibrosis from a routine, H&E-stained liver biopsy? 💡 This is the transformative potential of LiverExplore,* our suite of deep-learning models trained to exhaustively predict and quantify features of the liver microarchitecture – including fibrosis – directly from digitized H&E-stained liver biopsies. 📃 Read about LiverExplore’s ability to quantify fibrosis composition beyond the categorical CRN stages in our latest blog. LiverExplore and AIM-MASH+ are For Research Use Only. Not for use in diagnostic procedures. #PathAI #DigitalPathology #AI #ML #ComputerVision #ComputationalPathology #AASLD #Liver #Fibrosis