New Visual Search Engine Launches to Help Doctors Identify Complex Tumours

2026-09-13 biotech

Berlin, Sunday 13 September 2026
Aignostics has launched a visual search engine allowing researchers to instantly match tissue sample screenshots against a database of 310,000 digital scans to identify rare tumours.

Accelerating Differential Diagnosis with Atlas 2

The newly introduced PathoSearch platform is designed to tackle the significant bottleneck in identifying complex and rare tumours, which can represent up to 20% of a pathologist’s daily workload [2]. Built on the Atlas 2 pathology foundation model—co-developed by Berlin-based Aignostics and the Mayo Clinic—the visual search engine allows researchers to retrieve morphologically similar, pre-diagnosed reference cases using a single screenshot of a Hematoxylin and Eosin (H&E) slide [2][4][5][6]. By delivering comparable, fully annotated cases in minutes, the tool aims to drastically reduce the time required to secure informed second opinions [2][4]. Dr. Simon Schallenberg, an associate professor and pathologist at Charité – Universitätsmedizin Berlin, noted that while traditional digital image sharing or physical slide shipping can take hours or even days, instant access to PathoSearch reference cases ensures pathologists are much better informed from the outset [1][4][5].

Database Scale and Workflow Integration

At launch, PathoSearch draws on a curated database containing more than 310,000 whole slide images from over 35,000 validated cases [1][2][4][6]. This repository covers more than 300 diagnostic entities across 27 organs, aligned with the World Health Organisation (WHO) Classification of Tumours [1][2][4][6]. Rather than relying solely on direct standalone access, Aignostics is pursuing a platform-embedded distribution strategy [2]. The search engine has seen its first live commercial integration on Techcyte’s Fusion® AP image management system, allowing pathologists to query the database directly within their existing digital workflows [1][2][4][5][6]. Techcyte and Aignostics are actively demonstrating this integrated functionality at the 38th European Congress of Pathology (ECP) in Stockholm, which runs from 12 to 16 September 2026 [6].

Industry Contradictions and Market Consolidation

While multiple international news outlets and partner announcements have detailed the early-access rollout of PathoSearch [1][2][4][5][6], some industry analysis has disputed the product’s existence under this specific name, claiming it is not officially listed in certain corporate portfolios or filings [3]. Nevertheless, the strategic positioning of Aignostics’ underlying technology reflects a rapidly expanding computational pathology sector [2][3]. Aignostics, established in 2020 as a spin-off from Charité Berlin, has secured approximately $55 million in total venture funding, including a €14 million Series A in 2022 led by Wellington Partners and a $34 million Series B in October 2024 led by ATHOS [3]. This funding supports its proprietary Atlas 2 model, which features 2 billion parameters and was trained on over 5 million slide images [3].

The High-Stakes Digital Pathology Landscape

The launch of image-based search tools comes amid intense consolidation and financial growth within the digital pathology market [3]. In May 2026, Roche agreed to acquire PathAI for up to $1.05 billion, consisting of $750 million upfront and $300 million in milestone payments [3]. Meanwhile, Tempus AI acquired Paige for $81.25 million in stock, gaining access to Paige’s dataset of nearly 7 million digitized pathology slides [3]. This consolidation is driven by a rapidly growing market; the global digital pathology sector was valued at approximately $1.3 billion in 2025 and is projected to reach up to $5.75 billion by 2034, representing a projected increase of ƒ{(5.75 - 1.3) / 1.3 * 100}% [3]. Within this market, the AI-in-pathology niche alone was valued at $166 million in 2025, with an annual growth rate of roughly 24% forecast through 2034 [3].

Regulatory Boundaries and Scaling Plans

To manage regulatory exposure during this initial rollout, Aignostics has designated PathoSearch strictly for “Research Use Only” (RUO), meaning it is not currently cleared or authorised for direct clinical diagnostic procedures [2][4][5][6]. This free, early-access phase allows the company to gather vital pathologist feedback and validate performance patterns before formal commercialisation [2][5]. Looking ahead, Aignostics intends to scale the PathoSearch reference database to exceed one million whole slide images by the time of its general availability, with the ultimate goal of achieving comprehensive coverage of all WHO-classified malignant tumours [1][2][4][5].

Sources & Ecosystem Partners

AI pathology clinical diagnostics