Published January 26, 2026

How AI Detects Breast Cancer on Mammograms Study Results

How AI Helps Detect Breast Cancer on Mammograms: Results of a Russian Study

The L7 Center conducted an independent study of the «Celsus» algorithm on mammography images – the system demonstrated high accuracy in detecting breast pathologies.

Medicine
Event Source: Celsus Reading Time: 2 – 3 minutes

Medical algorithms are gradually becoming part of the diagnostic process, but before using them in clinical settings, their effectiveness must be verified. The L7 Center conducted an independent study of the artificial intelligence system from the company «Celsus», which analyzes mammography images.

What Was Studied

We are discussing an algorithm that assists doctors in detecting breast pathologies using X-rays – specifically, mammograms. Simply put, the system analyzes the image and highlights areas that could indicate a disease.

The L7 Center used a set of mammography examinations for verification and processed them using the algorithm. The task was straightforward: to evaluate how accurately the system identifies problematic areas and whether it misses anything important.

What the Results Showed

The results proved encouraging. The algorithm showed high sensitivity – meaning it accurately recognizes pathology when it is present. Concurrently, the system demonstrated acceptable specificity, indicating it does not produce too many false positives by identifying a problem where none exists.

For medical algorithms, this balance is crucial. If the system generates too many false alarms, doctors will cease to trust it. If it misses actual cases, patients will suffer negative consequences.

Why This Is Needed in Practice

Mammography is a mass screening examination. Radiologists view hundreds of images, and human attention is not limitless. AI can function as a «second pair of eyes» here: it highlights suspicious areas, and the doctor makes the final decision.

This is particularly useful in regions with a shortage of specialists or in large centers with a high volume of patients. The algorithm does not replace the doctor but helps reduce the workload and decrease the likelihood that something important will go unnoticed.

What's Next

The study by the L7 Center is an important step toward the wider use of such systems. Independent verification demonstrates that the technology works on real data, not just under idealized laboratory conditions.

Now the question is how quickly these algorithms will be integrated into routine practice. This requires not only technical readiness but also understanding from medical institutions, staff training, and integration with existing systems.

However, the direction is clear: AI in medical diagnostics is not science fiction but a tool that can help doctors work more accurately and faster right now.

#applied analysis #research review #neural networks #computer vision #biology #society #ai in medicine #ai reliability
Original Title: Опубликованы результаты исследования Центра L7 по применению ИИ в маммографии от компании Цельс
Publication Date: Jan 26, 2026
Celsus celsus.ai A Russian company developing AI solutions for healthcare and medical diagnostics.
Previous Article MareNostrum 5 Gets an AI Boost: New Contract to Expand the European Supercomputer Next Article How Dropbox Uses the Cursor AI Editor for Code Rewriting

From Source to Analysis

How This Text Was Created

This material is not a direct retelling of the original publication. First, the news item itself was selected as an event important for understanding AI development. Then a processing framework was set: what needs clarification, what context to add, and where to place emphasis. This allowed us to turn a single announcement or update into a coherent and meaningful analysis.

Neural Networks Involved in the Process

We openly show which models were used at different stages of processing. Each performed its own role — analyzing the source, rewriting, fact-checking, and visual interpretation. This approach maintains transparency and clearly demonstrates how technologies participated in creating the material.

1.
Claude Sonnet 4.5 Anthropic Analyzing the Original Publication and Writing the Text The neural network studies the original material and generates a coherent text

1. Analyzing the Original Publication and Writing the Text

The neural network studies the original material and generates a coherent text

Claude Sonnet 4.5 Anthropic
2.
Gemini 3 Pro Preview Google DeepMind step.translate-en.title

2. step.translate-en.title

Gemini 3 Pro Preview Google DeepMind
3.
Gemini 2.5 Flash Google DeepMind Text Review and Editing Correction of errors, inaccuracies, and ambiguous phrasing

3. Text Review and Editing

Correction of errors, inaccuracies, and ambiguous phrasing

Gemini 2.5 Flash Google DeepMind
4.
DeepSeek-V3.2 DeepSeek Preparing the Illustration Description Generating a textual prompt for the visual model

4. Preparing the Illustration Description

Generating a textual prompt for the visual model

DeepSeek-V3.2 DeepSeek
5.
FLUX.2 Pro Black Forest Labs Creating the Illustration Generating an image based on the prepared prompt

5. Creating the Illustration

Generating an image based on the prepared prompt

FLUX.2 Pro Black Forest Labs

Related Publications

You May Also Like

Explore Other Events

Events are only part of the bigger picture. These materials help you see more broadly: the context, the consequences, and the ideas behind the news.

Want to know about new
experiments first?

Subscribe to our Telegram channel — we share all the latest
and exciting updates from NeuraBooks.

Subscribe