Two questions get in the way of a diagnosis often enough to matter. The familiar one: "I've seen a pattern like this before — where?” And the one that stops you: “I've never seen anything quite like this – what could it be?” In conversations with collaborating pathologists, estimates of how often a case raises one of these questions range between 5% and 20%, depending on subspecialty and experience. In a busy pathology department, that translates to a daily event, not an occasional one.
Pathologists' work is getting more demanding. The WHO Classification of Tumours, also known as the "blue books," is one visible measure of this. The current edition runs to fourteen volumes, including volumes that did not exist in earlier editions; each edition has added new entities and more molecular criteria within each definition.¹ Telling those entities apart requires ever-finer subspecialization. Pathology reports have grown as a result: a study of more than half a million reports revealed that report length has increased over time, even as the number of cases has fallen.² There is simply more to know, and more to record, for every diagnosis.
At the same time, the pathology workforce is aging faster than it is being replaced. Close to a third of pathologists are already over 60 and expected to retire within the decade.³ Measured against rising cancer incidence, the workload per pathologist has climbed sharply: by one estimate, more than 40% in a single decade.⁴ Subspecialty expertise cannot be staffed everywhere. The result is more complexity and less time: often only minutes to evaluate, diagnose, and sign out a case.
The first read carries outsized weight: it decides which immunohistochemistry panel to order, and whether the answer comes back in one round or three. A wrong start doesn't just cost time — it compounds, since every subsequent step inherits the initial error. Even specialists can struggle: at one tertiary sarcoma center, second-opinion review changed the initial diagnosis in 38% of cases; in one quarter of cases, this change altered how the patient was managed.⁵
The available references for these cases are largely what they have always been. A textbook, if the right one is current and on the shelf. Digital reference platforms, if there is already a diagnosis to look up. An external consultation, which takes hours when images are shared digitally and days when glass is shipped, and which at major cancer centers runs from several hundred to several thousand dollars per case. Each of these options requires pathologists to translate a complex visual pattern into words. Meanwhile, the original image sits in front of us, even in labs that are not fully digitized. What pathologists lack is a way to search using the image, rather than words.
This is why Aignostics built PathoSearch. Instead of starting a query based on words, PathoSearch lets you begin with the image itself: it is a lookup application that searches by morphology and shows real, diagnosed reference cases drawn from curated archives at world-class centers. Picture a tumor made of nests of large cells with granular pink cytoplasm, divided by fine vessels and falling apart in the middle. A distinctive look, but not one every generalist will recognize immediately. It could be an alveolar soft part sarcoma: rare, mimicking paraganglioma or metastatic renal cell carcinoma, and prone to metastasize, sometimes already at presentation, sometimes decades later. Displayed beside confirmed cases with that morphology, this possibility surfaces in seconds. PathoSearch provides a direct route to a diagnosis instead of a long walk down the differential.
This is pattern recognition and photographic memory on a massive scale, assembled from some of the highest-quality archives in the world and accessible with a few easy clicks. Not a replacement for the skill that defines pathology, but an extension of it. PathoSearch surfaces real evidence you can open, compare, and challenge rather than providing a single answer from a black box. It is augmented intelligence: a tool to sharpen your judgment, not replace it, for cases where recognition is hardest. The pathologist remains the pilot; the final call is always yours.
PathoSearch is now available as a free early access via our waitlist. If the challenges described here sound familiar, give PathoSearch a try and tell us what you find. If you work in Techcyte, PathoSearch is available there too: your Techcyte contact can set up a demo.
References
- WHO Classification of Tumours (WHO Blue Books). International Agency for Research on Cancer. Accessed August 7, 2026. Link.
- Bonert M, et al. Evolution of anatomic pathology workload from 2011 to 2019 assessed in a regional hospital laboratory via 574,093 pathology reports. PLoS One. 2021;16(6):e0253876. doi:10.1371/journal.pone.0253876. Link.
- Gassenmaier V, et al. Understanding the pathology workforce: motivations, job satisfaction, and training perspectives in Germany. Virchows Arch. Published online October 2025. doi:10.1007/s00428-025-04277-1. Link.
- Metter DM, Colgan TJ, Leung ST, Timmons CF, Park JY. Trends in the US and Canadian pathologist workforces from 2007 to 2017. JAMA Netw Open. 2019;2(5):e194337. doi:10.1001/jamanetworkopen.2019.4337. Link.
- Rupani A, Hallin M, Jones RL, Fisher C, Thway K. Diagnostic differences in expert second-opinion consultation cases at a tertiary sarcoma center. Sarcoma. 2020;2020:9810170. doi:10.1155/2020/9810170. Link.
Author Bio
Julika Ribbat-Idel, MD, is Principal Pathologist at Aignostics in Berlin, where she helped build PathoSearch. Before joining Aignostics, she spent more than 10 years in academic pathology and patient care in Northern Germany as a board-certified pathologist. She holds a Habilitation and venia legendi in pathology from Lübeck university, where she continues to serve as a lecturer and examiner for medical board examinations and doctoral theses.