Fujitsu unveils ‘explainable AI’ for use in genomic medicine and cancer treatment planning: Fujitsu Global

The newly developed technology takes on challenges in the field of genomic medicine, demonstrating world-leading accuracy against key benchmarks for cancer typing.

Fujitsu limited

Kawasaki, May 9, 2024

Fujitsu today announced the development of explainable artificial intelligence technology that automatically draws on data in multiple formats, including text, images and numerical data, to create knowledge graphs that will help users more easily extract meaning from data sets at a glance. Large scale with high precision for areas including medicine.

To confirm the effectiveness of this technology, Fujitsu tested it on several key benchmarks in the medical field, including lung cancer type classification and survival prediction of breast cancer patients. These tests confirmed that Fujitsu’s technology can accurately support the identification of two main types of lung cancer, for example, illuminating the factors behind pathological classification based on key visual cues.

Fujitsu has also developed technology to extract and train algorithms on the different features of images with completely different representations of objects and to make very precise judgments. It is envisioned that this technology can be applied to train AI to support highly accurate assessments from pathological images for which sufficient training data cannot be prepared.

In the future, Fujitsu will continue to develop these multimodal technologies. ⁽¹⁾ For general use in a variety of different fields and disciplines. By the end of fiscal 2024, Fujitsu also plans to offer the newly developed technologies through the Fujitsu Research Portal. ⁽²⁾an environment that gives users the ability to quickly test Fujitsu’s advanced technologies.

About newly developed technology

Fujitsu has been conducting research and development on multimodal technologies that handle data from different fields and in a variety of formats, and has developed the following two AI technologies that combine multiple different data formats and images with completely different object representations to train algorithms and make conclusions from various perspectives.

1. Artificial intelligence technology that is trained by combining image data with completely different drawing methods, such as line drawings and photographs.

Fujitsu has developed a technology that combines data from images in which objects are drawn in completely different ways, such as images, line drawings, illustrations and photographs, to learn to accurately differentiate images (for example, to determine which subject is the subject). . With this method, the unique and common feature values ​​of each image type are extracted and used for training, and these unique or common feature values ​​are used for differentiation. This way, even when objects are drawn in different ways, the combination of multiple data types can be used to train the algorithm to make appropriate decisions.

Fujitsu evaluated this technology using three standard benchmarks in this field of research (PACS, Office-Home, DomainNet), which consist of data sets of multiple image types, including art, manga images, and photographs. As a result of comparative testing, Fujitsu confirmed that with the new technology the accuracy of object identification could be improved by approximately 2% compared to conventional technologies, where objects were identified using image data with completely different drawing methods. . This result was recorded at ICLR 2024 of the renowned conference The International Conference on Learning Representations (ICLR) ⁽³⁾and Fujitsu has presented it on May 8, 2024.

2. Explainable AI technology that integrates data from different formats and transforms it into a common knowledge graph for training.

To integrate different data formats, such as not only images but also text and images, Fujitsu has developed a new technology that converts different data into a common knowledge graph regardless of the original format by applying the above technology. This is automatically integrated with AI to create a large-scale integrated knowledge graph, which can be used to allow AI to make decisions in an explainable way.

By applying this technology to the following medical fields, Fujitsu anticipates results that could surpass the performance of conventional technologies.

In combination with newly developed technology to extract and train algorithms on the various features of images with completely different object representations, the technology is also expected to improve the accuracy of identifying pathological images for which sufficient image data cannot be prepared. training.

1) Classification of lung cancer

Treatments are currently being introduced for different types of lung cancer, such as adenocarcinoma. ⁽⁴⁾ and squamous cell carcinoma ⁽⁵⁾And accurate classification of cancer types is important to ensure correct treatment. In the past, a doctor would visually query multiple pieces of information and perform a thorough medical examination. However, the new technology offers the potential to significantly speed up this process, using AI to automatically integrate pathological images and genomic information (information about copy number abnormalities) from lung cancer patients to identify cancer types. As a result, when evaluated using data from the Cancer Genome Atlas (TCGA), a global standard benchmark, the technology achieved the world’s highest accuracy of 92.1% for lung cancer typing, compared to the previous highest accuracy of 87.1%. For these types of classification, the basis of judgment can be shown by returning to the pathological image data.

2) Determine the survival prediction of patients with breast cancer.

Being able to accurately predict the length of survival for each treatment when a person chooses a treatment method increases the likelihood that the appropriate treatment will be chosen. In this study, by automatically integrating and reviewing breast cancer patient image data as well as RNA data ⁽⁶⁾ and healthcare data using AI, Fujitsu evaluated them using reference data from The Cancer Genome Atlas (TCGA). In the task of predicting the survival time of breast cancer patients, Fujitsu’s technology achieved 71.8%, compared to the previous best accuracy of 66.8%. Imaging data can be used to provide evidence in predicting these survival times.

Future plans

The technology will be offered to users through the Fujitsu Research Portal in fiscal 2024.

In 2023, Fujitsu began collaborating with the Barcelona Supercomputing Center – Centro Nacional de Supercomputación (BSC-CNS) ⁽⁷⁾, the national supercomputing center of Spain, a leading research organization in the field of personalized medicine. The AI ​​technology developed using this multi-modal technology will also be used in joint research with the Barcelona Supercomputing Center to further improve accuracy and gain global recognition. Fujitsu will continue to develop this technology with a view to using it not only in the medical field, but also in various fields, such as data center failure prediction and fraud detection.

---

related links

Fujitsu’s commitment to the Sustainable Development Goals (SDGs)

Press contacts

---

All company or product names mentioned herein are trademarks or registered trademarks of their respective owners. The information provided in this press release is accurate at the time of publication and is subject to change without notice.

Date: May 9, 2024

City: Kawasaki, Japan

Company: Fujitsu limited