By Sumaiya
Google Health has joined hands with iCAD for creating an AI breast cancer screening tool
A commercial provider of medical technology has licensed Google Health’s AI breast cancer screening model, opening the door for the system developed by researchers to undergo its first clinical trials. The company Google Health has made a partnership with iCAD from New Hampshire-based, manufacturer of equipment for radiation therapy and cancer detection.
The business will start incorporating Google AI’s computer vision system, which is intended to spot breast cancer in mammography scans, to create a service that will eventually be sold to healthcare professionals. As per the agreement, the partnership with iCAD would provide infrastructure for safe data storage using Google Cloud services. iCAD has agreed to license Google’s breast cancer screening model for five years, according to an iCAD representative.
Following the completion of the commercial product, there will be a regulatory approval procedure. The iCAD cloud product is anticipated to be available by the end of 2023, while the AI solution with Google AI is anticipated to be available in the first half of 2024, albeit these dates are only estimates and are subject to change.
For years, Google Health has been developing an AI system to help physicians more correctly diagnose breast cancer using mammography scans. A group of computer scientists from Google Health, DeepMind, Verily Sciences, and colleges and universities in the UK and the US, including North-western University, Imperial College London, and the University of Cambridge, published a paper in Nature in 2020 asserting that AI could outperform trained radiologists in identifying cancerous breast tissue.
In comparison to six radiologists, the system apparently exhibited lower rates of false positives and false negatives. It was advertised as a means to cut down on pointless patient follow-ups, enabling doctors to give priority to women who are most at risk for developing the condition and accelerating diagnosis. Google AI could be a tough competition for other AI’s in the healthcare industry once Google starts functioning its AI in full swing.
According to Google Health, the model was trained using datasets of mammograms from more than 15,000 US women and 76,000 women in the UK. In studies with the experts, it resulted in a 5.7 percent reduction in false positives in the US and a 1.2 percent reduction in the UK, as well as a 9.4 percent reduction in false negatives in the US and a 2.7% drop in the UK.
Medical professionals and academics have previously criticized the firm for refusing to make the model’s code available so that others may duplicate and verify the findings months after the paper was published. The code is currently being distributed as a commercial license to businesses that can use it in clinical settings for actual patients.
Greg Corrado, the head of Google Health’s AI division, said in a statement that “the company’s AI technology may be used to improve healthcare availability, accessibility, accuracy, and diagnosis by Google Health”.
However, bringing about a change of this magnitude will only be feasible if they collaborate closely with partners that share our commitment to innovation and who have a long history of being among the first to introduce new ideas into the market. Google Health’s partnership with iCAD to improve breast cancer screening globally is a perfect example of two organizations using their expertise, technology resources, and resources to improve the health outcomes of people and communities.
Aside from its time spent working with iCad, Google has essentially been enhancing its healthcare domain. Diagnosis by Google Health is boosting health initiatives across its organization, from streamlining EHR operations to expand access to accurate health data and create new algorithms to identify disorders. These are some of the initiatives the tech juggernaut is working on.
The Advanced Technology and the Projects division of Google is creating a tiny computing component called Tag. According to Nicholas Gillian, machine learning head at Google’s Advanced Technology and Projects division, their objective is to develop a wearable device that can replicate the same kind of machine learning analysis that takes place in labs.
The EHR platform from Meditech now uses Care Studio, a Google Health EHR search tool. The alliance seeks to give Meditech users access to Care Studio’s data harmonization, search, and summarising capabilities.
By storing and processing data locally without connectivity, Google’s Fast Healthcare Interoperability Resources software development kit enables healthcare providers to run Android apps offline. The programme is a part of Google’s effort to use mobile technology to democratize access to high-quality healthcare.
