Pattern Unveils Adaptive Example Selection™ (AES™) Framework to Revolutionize Medical AI Interpretability

Innovative framework enhances clinician trust and transparency in AI-driven healthcare diagnostics, driving broader adoption of AI across clinical practices

/EIN News/ -- REDMOND, Wash., March 12, 2025 (GLOBE NEWSWIRE) -- Pattern Computer®, Inc. (“Pattern” or “the Company”), a leader in cutting-edge explainable AI (XAI) research, has introduced a transformative new framework, Adaptive Example Selection™ (AES™), designed to revolutionize the way clinicians interact with and trust AI systems in high-stakes medical environments. This framework, entitled “Adaptive Example Selection: Prototype-Based Explainability for Interpretable Mitosis Detection,” has been published in a new study on bioRxiv, an online archive and distribution service for unpublished preprints in the life sciences. The framework significantly enhances the interpretability of AI, particularly in complex diagnostic tasks such as mitosis detection for cancer grading and prognosis. With AES, AI models become more transparent, adaptable, and clinically relevant, opening new doors for collaboration between clinicians and AI systems.

AI-driven solutions in healthcare hold immense promise, but one of the most significant barriers to their widespread adoption has been the “black-box” nature of deep learning models. Pattern has taken a step towards overcoming this challenge by developing AES, designed to improve diagnostic accuracy and render the decision-making process transparent and interpretable. AES presents clinicians with prototype-based, real-world examples that mirror newly classified cases, allowing them to directly compare AI predictions to known, annotated data.

The AES framework leverages Decision Boundary-based Analysis, a novel method that expands and fits the model’s belief function to a radial basis function. This enhances the transparency of decision boundaries, helping clinicians understand the model’s predictions with greater clarity. AES is fully customizable, giving healthcare professionals the ability to tailor decision thresholds and select prototype examples, ensuring the system meets the specific needs of their practice.

“Our vision at Pattern is to make AI not just accurate, but also interpretable and adaptable to real-world applications,” said Mark Anderson, Chair and CEO. “AES empowers clinicians with the transparency they need to trust AI in critical decision-making, allowing them to use these tools with confidence in diverse medical environments. This is an important key to unlocking AI’s full potential in healthcare.”

Key Benefits of AES:

• Transparency in decision-making: AES transforms black-box models into clear, understandable tools by providing visual, interpretable explanations.
• Adaptability across domains: While initially designed for mitosis detection, AES is adaptable to a wide range of diagnostic challenges, including tumor detection, organ classification, and rare disease identification.
• Increased clinician trust and collaboration: AES fosters a more collaborative relationship between clinicians and AI, enabling a deeper understanding of model behavior and reducing reliance on blind trust.
• Customizable user interface: Clinicians can adjust decision-making parameters to align with their diagnostic needs, making AES a highly flexible solution.
• Scalability and generalizability: AES’s principles extend beyond healthcare, offering a framework for improving interpretability across numerous industries where black-box deep learning models are used.
  

Anderson concluded, “Pattern’s continued research and development in XAI are setting new standards for transparency and trust in AI systems. AES is just one example of how the Company is paving the way for more intuitive, interpretable AI applications that empower professionals across sectors to make more informed, data-driven decisions.”

About Pattern
Pattern Computer, Inc. uses its Pattern Discovery Engine™ to solve the most important and intractable problems in business and medicine. These proprietary mathematical techniques in advanced AI can find complex patterns in very-high-order data that have eluded detection by much larger systems. As the Company applies its computational platform to the challenging fields of drug discovery and diagnostics, it is also making major Pattern Discoveries for partners in other sectors, including extended biotech, materials science, aerospace manufacturing quality control, veterinary medicine, air traffic operations, and energy services. See www.patterncomputer.com.

CONTACT: Laura Guerrant-Oiye (808) 960-2642 – laura@patterncomputer.com

The foregoing contains statements about Pattern Computer’s future that are not statements of historical fact. These statements are “forward-looking statements” for purposes of applicable securities laws and are based on current information and/or management’s good faith belief as to future events. The words “believe,” “expect,” “anticipate,” “project,” “should,” “could,” “will,” and similar expressions signify forward-looking statements. Forward-looking statements should not be read as a guarantee of future performance. By their nature, forward-looking statements involve inherent risk and uncertainties, which change over time, and actual performance could differ materially from that anticipated by any forward-looking statements. Pattern Computer undertakes no obligation to update or revise any forward-looking statement.

Copyright © 2025 Pattern Computer Inc. All Rights Reserved. Pattern Computer, Inc., Pattern Discovery Engine, PatternBio, TrueXAI, and ProSpectral are trademarks of Pattern Computer Inc. or its subsidiaries. Other trademarks may be trademarks of their respective owners.