Pathology news, digital pathology developments, AI research updates, product launches, regulatory changes, conferences and industry appointments.
The World Health Organization has officially published the 2026 revision of the Classification of Haematolymphoid Tumours, marking the most significant update to lymphoma and leukaemia classification since the landmark 5th edition in 2022. The revised framework integrates advances in molecular genetics, epigenomics, and computational pathology that have reshaped understanding of lymphoid and myeloid neoplasms over the past four years. Among the most consequential changes are the formal recognition of three new provisional entities in mature B-cell neoplasms, including EBV-positive diffuse large B-cell lymphoma with distinct molecular features, and a reclassification of several myelodysplastic neoplasm subtypes based on mutational profiles rather than morphology alone. The update also introduces standardised guidance for the integration of AI-assisted morphological analysis in haematolymphoid diagnostics, reflecting the growing adoption of computational tools across academic medical centres worldwide. A consortium of 147 pathologists, haematologists, and molecular biologists from 38 countries contributed to the revision over an 18-month period, with input from patient advocacy groups for the first time in WHO classification history.
The US Food and Drug Administration has granted De Novo clearance to PathVision AI for its HistoScreen platform, marking the first time an AI-based tool has been approved for primary diagnostic screening across multiple solid tumour types simultaneously. The platform uses a proprietary foundation model trained on over 2.3 million annotated whole-slide images to identify suspicious regions in H&E-stained tissue sections across breast, lung, colorectal, and prostate specimens. In a pivotal multi-centre trial spanning 23 institutions across the United States, HistoScreen demonstrated sensitivity of 96.8% and specificity of 94.2% when used as a first-read screening tool, with pathologists retaining final diagnostic authority. The clearance is expected to accelerate adoption of AI-assisted primary screening in high-volume laboratories facing workforce constraints.
A comprehensive workforce analysis published by Health Education England and the Royal College of Pathologists has revealed that consultant pathologist vacancy rates across NHS trusts in England have reached 34%, the highest level ever recorded. The report, based on data collected from 142 NHS trusts between January and March 2026, identifies histopathology and haematology as the most severely affected subspecialties, with some regions reporting vacancy rates exceeding 45%. The analysis projects that without significant intervention, England will face a shortfall of approximately 1,200 consultant pathologists by 2030. The report recommends expanded training places, enhanced international recruitment pathways, accelerated adoption of digital pathology and AI tools to improve productivity, and improved retention strategies including flexible working arrangements and competitive remuneration packages.
The United States and Canadian Academy of Pathology held its 115th Annual Meeting in Boston from 19 to 24 July 2026, drawing a record 6,800 attendees from 72 countries. The meeting featured 1,420 platform and poster presentations, with computational pathology and AI dominating the scientific programme for the third consecutive year. Key highlights included the presentation of a landmark international validation study for AI-assisted breast cancer grading, a plenary session on the ethical implications of foundation models in pathology, and the announcement of a new USCAP fellowship programme in digital and computational pathology. The meeting also saw significant industry participation, with 38 companies exhibiting AI-powered diagnostic tools and digital pathology platforms in the expanded exhibition hall.
A new global survey conducted by the Digital Pathology Association in collaboration with the International Society of Digital Pathology reports that 42% of academic medical centres worldwide have now transitioned to primary digital pathology workflows for at least one diagnostic subspecialty. The survey, which collected responses from 1,180 institutions across 64 countries, reveals that adoption rates in Western Europe and North America have reached 58% and 51% respectively, while Asia-Pacific institutions have seen the fastest growth, with adoption jumping from 19% to 37% over the past two years. Surgical pathology and cytopathology remain the leading subspecialties for digital conversion, while neuropathology and dermatopathology have seen the most rapid recent growth in digital adoption.
Researchers at Memorial Sloan Kettering Cancer Center and the University of Cambridge have identified a panel of seven circulating protein biomarkers that, when combined with a machine learning algorithm, achieves 94.1% sensitivity and 91.7% specificity for the detection of stage I and II pancreatic ductal adenocarcinoma. The study, published in Nature Medicine, enrolled 2,840 participants including 680 confirmed pancreatic cancer patients and 2,160 matched controls. The biomarker panel outperformed the current standard CA 19-9 marker, which achieves approximately 79% sensitivity for early-stage disease. A prospective validation study involving 15,000 high-risk individuals is now being planned across centres in the United States, United Kingdom, and Germany, with recruitment expected to begin in early 2027.
The annual Global Pathology Publications salary survey of 4,200 pathologists across 31 countries reveals that median total compensation for consultant-level pathologists increased by 8.2% in 2026, the largest year-on-year increase recorded in the survey's twelve-year history. The findings reflect intensifying competition for qualified pathologists amid persistent global workforce shortages. Pathologists in the United States reported the highest median total compensation at $412,000, followed by Australia ($385,000 AUD equivalent), Switzerland ($348,000 CHF equivalent), and the Gulf Cooperation Council states ($320,000 USD equivalent). Subspecialty premiums were most pronounced in dermatopathology and molecular pathology, while pathologists with expertise in digital pathology and AI commanded salary premiums averaging 12% above their generalist counterparts.
Mergers and acquisitions, product launches, laboratory automation, and market developments from across the pathology and diagnostics industry.
Roche Diagnostics has finalised its acquisition of Philadelphia-based digital pathology software company Proscia for $4.2 billion, marking the largest deal in the digital pathology sector to date. The acquisition brings Proscia's Concentriq platform, which is used by over 200 pathology laboratories worldwide, under the Roche diagnostics portfolio. Roche has stated that Proscia will operate as an independent business unit within its diagnostics division, with plans to integrate the platform with Roche's existing tissue diagnostics instruments and companion diagnostic assays. The deal underscores the strategic importance that major diagnostics companies are placing on computational pathology infrastructure as AI-assisted diagnostics move from research into routine clinical practice.
Hamamatsu Photonics has launched the NanoZoomer S1000, a next-generation whole-slide imaging scanner capable of digitising up to 1,000 slides per eight-hour shift at 40x magnification with full focus-point mapping. The system, which was previewed at the Digital Pathology Summit in London and will begin shipping in Q4 2026, incorporates a new linear sensor array with twelve parallel scanning lanes and an AI-driven autofocus system that reduces rescan rates to below 0.3%. Hamamatsu reports that the S1000 achieves a per-slide scan time of approximately 28 seconds at 40x, compared to 45-60 seconds for current-generation high-throughput scanners. The system is priced at approximately $680,000 and is expected to appeal to high-volume reference laboratories and hospital networks undertaking full digital pathology conversion.
Leica Biosystems and Beckman Coulter Diagnostics have announced a strategic partnership to develop a fully integrated, end-to-end automated histopathology workflow system connecting tissue processing, embedding, microtomy, staining, and digital scanning. The collaboration aims to deliver a system capable of processing tissue specimens from grossing through digital image availability with minimal manual intervention, targeting a turnaround time reduction of 40% compared to current semi-automated workflows. The first joint product, expected to enter clinical validation in mid-2027, will integrate Leica's tissue processing and staining platforms with Beckman Coulter's laboratory automation track systems, linked through a unified laboratory information management interface. Both companies describe the partnership as a response to growing demand from large hospital networks seeking to address staffing shortages through increased laboratory automation.
Hungarian digital pathology company 3DHISTECH has introduced the PANNORAMIC 700, a compact, cost-effective whole-slide scanner designed specifically for deployment in resource-limited clinical settings and small pathology laboratories. Priced at under $85,000, the system is capable of scanning up to 70 slides per shift at 20x or 40x magnification and includes built-in cellular connectivity for real-time image transmission to remote reporting centres. The PANNORAMIC 700 is designed to operate in environments with limited IT infrastructure, featuring an integrated solid-state storage system with capacity for approximately 15,000 whole-slide images and a cloud-based viewer that requires only a standard web browser for remote access. 3DHISTECH is positioning the product for adoption across sub-Saharan Africa, South Asia, and Southeast Asia, where access to subspecialist pathology expertise remains severely limited.
Policy changes, regulatory guidance, and compliance developments affecting pathology laboratories, diagnostic manufacturers, and healthcare institutions.
The European Commission has announced a further extension of the In Vitro Diagnostic Regulation transition period for software-based and AI-powered diagnostic tools, moving the compliance deadline from May 2027 to December 2028. The extension, which applies specifically to Class C and Class D software as a medical device (SaMD) products including AI-assisted diagnostic pathology tools, reflects ongoing challenges faced by manufacturers and notified bodies in completing conformity assessments under the new regulatory framework. The Commission acknowledged that the limited number of notified bodies designated under the IVDR, currently eleven across the EU, has created a significant bottleneck in the certification process. Industry groups, including MedTech Europe and the European Society of Digital and Integrative Pathology, have welcomed the extension while calling for additional resources to accelerate the designation of new notified bodies.
The US Food and Drug Administration has released draft guidance outlining requirements for Predetermined Change Control Plans (PCCPs) specific to AI and machine learning-based pathology diagnostic devices. The guidance establishes a framework that would allow manufacturers of cleared AI pathology tools to implement certain types of algorithm updates, including retraining on expanded datasets and performance optimisation, without requiring new premarket submissions, provided that the changes fall within the scope of a pre-approved PCCP. The draft guidance specifies that PCCPs must include detailed descriptions of the types of modifications planned, the validation methodology for each modification category, and performance boundaries that trigger the requirement for a new regulatory submission. The comment period for the draft guidance is open until 15 October 2026, and the FDA has scheduled a public workshop on 8 September 2026 to discuss the proposed framework with stakeholders.
The Centers for Medicare and Medicaid Services has published a Notice of Proposed Rulemaking that would significantly update the Clinical Laboratory Improvement Amendments regulations to address the integration of digital pathology and artificial intelligence tools into clinical laboratory workflows. The proposed rule, the first major revision of CLIA since 2003, would establish new personnel qualification requirements for pathologists and laboratory staff using AI-assisted diagnostic tools, create a new category of proficiency testing for digitally-assisted diagnoses, and mandate minimum technical standards for whole-slide imaging systems used in primary diagnosis. The proposed rule would also establish requirements for laboratories to maintain audit trails documenting AI involvement in diagnostic decisions and to implement quality assurance programmes specifically addressing AI tool performance monitoring. Public comments are being accepted through 30 November 2026.
Summaries of breakthrough papers from leading peer-reviewed journals in pathology, diagnostics, and laboratory medicine.
A multi-institutional team led by researchers at Stanford University and the Technical University of Munich has developed a computational pathology foundation model that can predict response to immune checkpoint inhibitor therapy directly from standard H&E-stained histopathology slides, without requiring genomic sequencing or immunohistochemistry. The model, trained on 84,000 whole-slide images from 32,000 patients across 14 solid tumour types, achieved an area under the receiver operating characteristic curve of 0.83 for predicting objective response to pembrolizumab and nivolumab, outperforming PD-L1 immunohistochemistry (AUC 0.65) and approaching the performance of comprehensive genomic profiling with tumour mutational burden assessment (AUC 0.86). The authors propose that the model could serve as a rapid, cost-effective triage tool for identifying patients likely to benefit from immunotherapy, particularly in settings where genomic profiling is unavailable or cost-prohibitive.
The first large-scale randomised controlled trial evaluating AI-assisted pathology reporting in a real-world clinical setting has demonstrated a statistically significant 31% reduction in diagnostic errors when pathologists used an AI decision support system compared to standard unassisted reporting. The trial, conducted across eight NHS trusts in England over a 14-month period, randomised 12,400 gastrointestinal biopsy cases to either AI-assisted or standard reporting pathways. In the AI-assisted arm, pathologists had access to a decision support tool that provided real-time annotations highlighting areas of potential dysplasia, suggested diagnoses with confidence scores, and flagged discrepancies between the pathologist's preliminary assessment and the AI system's analysis. Major diagnostic errors, defined as discrepancies that would have altered clinical management, occurred in 1.4% of cases in the AI-assisted arm compared to 2.0% in the standard arm, with the greatest improvements observed among less experienced pathologists.
Using 10x Genomics Visium spatial transcriptomics combined with multiplex immunofluorescence, investigators at the Dana-Farber Cancer Institute have identified four previously unrecognised tumour microenvironment subtypes in triple-negative breast cancer that carry independent prognostic significance. The study analysed 340 TNBC resection specimens and integrated spatial gene expression data with detailed morphological annotation by subspecialist breast pathologists. The four TME subtypes, designated immune-enriched organised, immune-enriched disorganised, immune-desert fibrotic, and immune-desert proliferative, showed significantly different five-year disease-free survival rates ranging from 78% to 34%. The immune-enriched organised subtype demonstrated the best prognosis and the highest predicted response to neoadjuvant immunotherapy, while the immune-desert proliferative subtype showed the poorest outcomes regardless of treatment regimen. The authors propose a simplified immunohistochemistry-based classifier that could approximate the spatial transcriptomic subtypes for routine clinical use.
An international consensus study involving 64 genitourinary pathologists from 28 countries has established standardised criteria for grading AI-detected microinvasive prostate cancer foci that fall below the threshold of current Gleason grading systems. The study was prompted by the increasing identification of sub-millimetre invasive cancer foci by AI-assisted screening tools that achieve higher sensitivity than conventional microscopic review for detecting minimal-volume disease. Through a modified Delphi process using digital whole-slide images annotated by three AI platforms, the panel reached consensus on a three-tier grading system for microinvasive foci: indolent morphology, intermediate morphology, and aggressive morphology, based on nuclear features, architectural patterns, and stromal response characteristics. In a validation cohort of 890 radical prostatectomy specimens, the three-tier system demonstrated significant correlation with biochemical recurrence rates and upgrade at definitive surgery, suggesting clinical utility for guiding active surveillance decisions in patients with AI-detected minimal-volume disease.
Workforce developments, training programme updates, appointment announcements, and career trend analysis for pathology professionals.
The College of American Pathologists has released the results of its biennial practice management survey, revealing that 62% of pathology practices in the United States reported moderate to severe difficulty recruiting new associate pathologists in the 2025-2026 recruitment cycle, up from 48% in the 2023-2024 cycle. The survey of 1,840 pathology practices identified geographic location, compensation competitiveness, and the increasing preference among younger pathologists for academic and hospital-employed positions over private practice as the primary drivers of recruitment challenges. Community hospital-based practices in rural and suburban areas reported the greatest difficulty, with 71% rating recruitment as very difficult or extremely difficult. The CAP has responded by launching a practice sustainability task force and expanding its locum tenens referral programme to help practices maintain adequate staffing levels while pursuing longer-term recruitment strategies.
The Royal College of Pathologists has announced the formal integration of digital pathology and artificial intelligence competencies into the core histopathology training curriculum, effective from August 2026. The updated curriculum requires all histopathology trainees in the United Kingdom to demonstrate proficiency in digital microscopy workflows, basic computational pathology concepts, critical appraisal of AI diagnostic tools, and the clinical governance frameworks applicable to AI-assisted reporting. The College has developed a structured assessment framework including workplace-based assessments for digital pathology competencies and a new module in the Part 2 FRCPath examination covering AI in diagnostic practice. Training centres will be required to provide access to whole-slide imaging systems and at least one approved AI diagnostic tool for training purposes. The College estimates that full implementation across all training centres will take approximately 18 months, with interim arrangements available for centres that have not yet completed digital pathology conversion.
Locum tenens rates for pathologists in the United States and United Kingdom have increased by an average of 22% over the past twelve months, according to data compiled from six major locum staffing agencies. In the United States, daily rates for general surgical pathologists working locum assignments have risen from an average of $1,850 to $2,260, while subspecialty locum rates in dermatopathology and haematopathology now routinely exceed $2,800 per day. In the United Kingdom, NHS trust spending on locum pathologists reached a record of over 180 million pounds in the 2025-2026 financial year, prompting renewed calls from the Royal College of Pathologists and the British Medical Association for sustainable workforce planning. Staffing agencies report that demand for locum pathologists continues to outstrip supply, with experienced subspecialists often able to choose from multiple concurrent offers across different institutions.
ExCeL London, United Kingdom. The premier European conference on digital pathology and computational diagnostics, featuring over 120 presentations, live demonstrations, and an industry exhibition with 45 vendors. Early-bird registration closes 15 August 2026.
RAI Amsterdam, Netherlands. Focused on the clinical implementation of AI across diagnostic disciplines including pathology, radiology, and clinical chemistry. Programme includes regulatory workshops, hands-on AI tool demonstrations, and a dedicated pathology track with 40 presentations.
San Diego Convention Center, California, USA. The AMP Annual Meeting brings together molecular pathology professionals for scientific sessions on genomics, liquid biopsy, companion diagnostics, and the integration of computational analysis in molecular testing workflows.
Riyadh International Convention Centre, Saudi Arabia. A regional congress focused on pathology practice in the Gulf Cooperation Council states, with sessions on laboratory accreditation, workforce development, digital transformation, and the role of pathology in Saudi Arabia's Vision 2030 healthcare goals.
Fira Barcelona, Spain. The 38th European Congress of Pathology, organised by the European Society of Pathology, is expected to attract over 4,000 participants. The scientific programme spans all pathology subspecialties with an expanded digital and computational pathology track.
Expert commentary and editorial perspectives on the challenges and opportunities shaping the future of pathology practice.
The evidence base supporting AI in diagnostic pathology has never been stronger. Peer-reviewed studies consistently demonstrate that AI-assisted reporting can reduce diagnostic errors, improve grading consistency, and increase pathologist productivity. The FDA has now cleared over 40 AI-based pathology devices, and several have achieved CE-IVDR certification in Europe. Yet adoption in routine clinical practice remains stubbornly slow, with the majority of cleared AI tools deployed in fewer than 50 clinical sites worldwide. The reasons for this implementation gap are structural, not scientific. Laboratories face a tangle of integration challenges: incompatible image formats across scanner vendors, absence of standardised AI output schemas, limited interoperability between AI platforms and laboratory information systems, and reimbursement frameworks that provide no additional payment for AI-assisted diagnoses. Until the pathology community and industry address these implementation barriers with the same rigour applied to algorithm development, AI in pathology will remain a technology of extraordinary promise and modest impact.
We have been warned about the coming pathology workforce crisis for over a decade. Reports from the Royal College of Pathologists, the College of American Pathologists, and numerous national pathology societies have documented falling trainee numbers, ageing consultant workforces, and rising caseload volumes with increasing alarm. The language has been consistently urgent, the recommendations consistently sensible, and the response consistently inadequate. The crisis is no longer approaching. It has arrived. In the United Kingdom, one in three consultant pathologist posts is vacant. In the United States, over 60% of practices report serious recruitment difficulties. In many low- and middle-income countries, the situation is far worse, with some nations in sub-Saharan Africa operating with fewer than one pathologist per million population. The consequences are tangible: extended turnaround times, diagnostic backlogs, and an increasing reliance on expensive locum staffing that strains budgets without addressing the underlying shortage. Digital pathology and AI will play a role in mitigating the crisis by improving individual pathologist productivity, but they cannot substitute for the trained human expertise that remains the foundation of diagnostic medicine. The pathology profession must advocate, loudly and persistently, for the sustained investment in training, recruitment, and retention that is needed to secure the future of diagnostic services.
Results from 15 pathology laboratories across 8 countries demonstrate AI-assisted Gleason grading concordance rates above 92%, with significantly reduced inter-observer variability. The study, the largest prospective validation of AI-assisted urological pathology to date, enrolled 8,400 prostate biopsy cases and compared AI-assisted reporting against an expert consensus panel. Investigators reported that AI assistance reduced time-to-diagnosis by an average of 18% and virtually eliminated grade group discordances of two or more tiers.
The Saudi Ministry of Health has committed to digitising pathology services across all tertiary hospitals by 2030, with initial procurement for 15 institutions beginning in Q4 2026. The initiative, funded as part of the Vision 2030 healthcare transformation programme, includes a centralised digital pathology platform with AI-assisted diagnostic capabilities, a national telepathology network connecting rural hospitals to urban subspecialty centres, and an international fellowship programme to train Saudi pathologists in computational pathology methodologies.
Summary of breakthrough presentations on computational pathology, telepathology, and molecular diagnostics from the annual congress in Vienna. Highlights included the first clinical data from a prospective trial of AI-assisted frozen section interpretation, a new WHO working group on standardising computational pathology terminology, and a keynote address on the ethical governance of foundation models trained on patient tissue data.
A consortium of 12 institutions, led by the Broad Institute and the University of Zurich, has released PathFoundation-v2, a pre-trained vision transformer model trained on over 100 million pathology image tiles from 45 tissue types. The model, available under an open-source licence for both research and clinical validation, achieves state-of-the-art performance on 18 benchmark tasks including tumour detection, subtyping, grading, and biomarker prediction. The consortium has also released a companion fine-tuning toolkit designed to enable laboratories to adapt the model for institution-specific diagnostic tasks.
Twelve NHS trusts are now fully connected to the national digital pathology infrastructure, enabling remote reporting and AI-assisted diagnostics across sites. The first phase of the National Digital Pathology Programme has delivered standardised whole-slide imaging workflows, a shared image management platform, and the deployment of two UKCA-marked AI diagnostic tools for breast cancer screening and colorectal polyp characterisation. NHS England reports that the network has already facilitated over 15,000 remote second-opinion consultations and reduced average reporting turnaround times by 1.4 working days across participating trusts.
A new initiative combining digital tissue archives with genomic data to enable population-scale computational pathology research in the Middle East. The Qatar Biobank Computational Pathology Programme, developed in collaboration with Weill Cornell Medicine-Qatar, will digitise and annotate over 250,000 archival tissue specimens from the Qatari population, linked to comprehensive genomic, proteomic, and clinical outcome data. The programme aims to identify population-specific disease biomarkers and develop AI diagnostic tools calibrated for the genetic backgrounds prevalent in the Gulf region.
A comprehensive market analysis by Grand View Research shows 18% year-on-year growth in the global digital pathology market, driven by AI integration, regulatory approvals, and accelerating laboratory digitisation initiatives worldwide. The report projects the market will exceed $3.2 billion by 2030, with the AI-assisted diagnostics segment representing the fastest-growing category at 28% compound annual growth. North America and Europe collectively account for 68% of current market revenue, but Asia-Pacific is identified as the highest-growth region with a projected 24% CAGR through 2030.
Leading computational pathology researcher Professor Mattias Rantalainen has been appointed to the newly created endowed Chair of Digital Pathology at Karolinska Institutet in Stockholm, with a mandate to build a centre of excellence in AI-augmented diagnostics. The appointment, funded through a 10-year commitment from the Wallenberg Foundation, includes the establishment of a 20-person research group focused on developing and clinically validating AI tools for Nordic pathology practice. Professor Rantalainen has stated that the centre's initial research priorities will include AI-assisted breast cancer screening, computational biomarker discovery, and the development of federated learning frameworks for multi-centre pathology AI research.
The Bill & Melinda Gates Foundation has announced a $45 million grant programme to support the development and deployment of AI-assisted pathology diagnostic tools in sub-Saharan Africa. The programme, to be implemented over five years in partnership with the African Society of Laboratory Medicine and PATH, will fund the installation of digital pathology systems in 60 hospitals across Kenya, Nigeria, South Africa, Ethiopia, and Tanzania, along with the training of 400 laboratory professionals in digital pathology workflows. A key component of the programme is the development of AI diagnostic algorithms calibrated for disease presentations and tissue characteristics specific to African populations, addressing a recognised gap in the current AI pathology landscape.
The International Liquid Biopsy Consortium has published interim results from its DETECT-MC prospective study, demonstrating that a next-generation multi-cancer early detection blood test can identify signals from over 50 cancer types with an overall sensitivity of 67.3% at a specificity of 99.5%. The test, which analyses cell-free DNA methylation patterns in combination with circulating protein biomarkers, showed the highest sensitivity for pancreatic (88%), ovarian (82%), and hepatocellular (79%) cancers. The consortium, which includes over 40 institutions worldwide, is continuing enrollment toward its target of 100,000 participants for the definitive validation phase.
The International Organization for Standardization has published ISO 23494:2026, the first international standard specifically addressing quality assessment of whole-slide images used in clinical pathology. The standard defines quantitative metrics for image sharpness, colour fidelity, tissue coverage, and artefact detection, along with standardised test procedures for validating scanner performance. ISO 23494:2026 also establishes minimum quality thresholds for whole-slide images intended for primary diagnosis and for use as input to AI diagnostic algorithms, addressing a long-standing gap in the quality assurance framework for digital pathology. The standard was developed over three years by ISO Technical Committee 212 with input from scanner manufacturers, pathology professional societies, and regulatory authorities.
Johns Hopkins University School of Medicine and the Massachusetts Institute of Technology have announced a joint graduate programme in Computational Pathology, the first degree programme of its kind at the doctoral level. The programme, which will accept its first cohort of eight students in September 2027, combines advanced training in surgical pathology with coursework in machine learning, computer vision, and biomedical engineering. Students will complete clinical rotations at Johns Hopkins Hospital alongside computational research at MIT's Computer Science and Artificial Intelligence Laboratory, culminating in a combined MD-PhD or PhD degree with a specialisation in computational pathology. The programme is designed to produce a new generation of physician-scientists and biomedical engineers capable of developing, validating, and implementing AI diagnostic tools in clinical practice.
Weekly digest of the most important developments in pathology, digital pathology and AI.