Deciphex Launches CipherX to Audit AI Pathology Reads

Deciphex has released CipherX, an artificial intelligence engine designed to make pathology diagnostics more transparent. The system aims to solve a common frustration in medical imaging: the inability to see why an algorithm flagged a specific cell cluster or tissue pattern.

By adding a human-readable layer on top of complex mathematical models, the platform lets doctors trace digital predictions back to physical features on a glass slide.

The company reports that this approach yields high reliability in routine casework, with detection algorithms showing a 99.85% negative predictive value for adenocarcinoma. This means the system is highly accurate in confirming the absence of the disease, reducing the need for unnecessary follow-up tests. The technology has been in active production since January 2026, serving as the core engine for both clinical diagnostics and pharmaceutical research.

Translating data into visible tissue

Standard AI models process images as dense clouds of numbers, which are difficult for humans to inspect. CipherX introduces a proprietary semantic layer that breaks these numbers down into discrete units called glyphs. These glyphs represent recurring structural elements that subspecialist pathologists have confirmed and named. By assembling these units into recognizable histological signatures, the system constructs higher-order spatial arrangements without requiring new model training.

This architecture creates a stable interface between the underlying neural networks and the clinical workflow. Because each output links directly to named structures that a reviewing pathologist can inspect and reject, the system remains auditable. If the underlying foundation model is updated or replaced, the validated clinical vocabulary established with laboratory customers remains unchanged, ensuring continuity in diagnostic standards.

Performance metrics in clinical labs

In routine operations, the engine supports three specific assistance tools. These focus on pre-analytical image quality assessment, complexity-based case triage, and post-authorization quality review. By operating outside the primary diagnostic pathway, the tools help manage workload and ensure data integrity before a human pathologist makes a final decision.

Beyond routine diagnostics, the platform supports digital tissue biomarker development for pharmaceutical research. This application allows researchers to identify potential drug targets with greater precision. The company notes that this dual focus on clinical care and research demonstrates the engine's versatility across different stages of the medical pipeline.

Regulatory needs and model stability

Regulatory bodies increasingly demand auditable systems that do not require downstream tools to be redeveloped when an underlying foundation model is updated. Deciphex states that its architecture addresses this need by decoupling the diagnostic vocabulary from the specific neural network used. This design choice mitigates the risk of regulatory non-compliance when AI models evolve, a common challenge in rapidly changing tech sectors.

According to Clinical Lab Products, the industry has focused on expanding dataset sizes over the past three years, but performance at the tile level has largely converged. The new system argues that the next step is not just bigger data, but better interpretation. By providing a stable layer that speaks in pathology terms, the platform aims to bridge the gap between raw computational power and clinical trust.