Israel’s DIA receives ڈالر 14 million to enhance AI-powered ultrasound analysis.

Israel-based AI health tech company DI Imaging Analysis, which uses deep learning and machine learning to automate ultrasound scan analysis, has cut off ڈالر 14 million in Series B round funding. ۔

Supporters of the growth era, which comes three years after the last DIA was raised, include new investors Alchemia Ventures, Downing Ventures, Icon Fund, Philips and XTX Ventures – existing investors. With CEO Ventures, Connecticut Innovations, Defta Partners, Mindset Ventures, and Dr. Shmuel Kabli. In total, it has taken in 25 25 million to date.

The latest financing will go beyond expanding its product range and new and expanded partnerships with Ultrasound Vendors, PACS / Healthcare IT Companies, Resellers and Distributors while continuing its presence in the three regional markets. Will

The health tech company sells AI-powered support software to physicians and healthcare professionals to help capture and analyze ultrasound imaging-a process that, when done manually, Visual interpretation of scanned data requires human skills. DIA therefore describes its AI technology as “extracting subjectivity from the process of manual and visual estimation performed today”.

It has trained AI to review ultrasound imaging to automatically focus on key details or identify abnormalities – offering a range of products targeting various clinical needs associated with ultrasound analysis, including the heart. Many focus on (where its software, for example, can be used to measure and analyze aspects such as ejection fractions. Help me).

It also has a product that uses ultrasound data to automate bladder volume.

The DIA claims that the way its AI software detects the boundaries of the human eye and identifies movement – presents it as a step beyond “subjective” human analysis in terms of speed and efficiency. Also adds.

“Our software tools are helpful tools for physicians to get the right picture and translate ultrasound data,” says Hilla Goldman Aslan, CEO and co-founder.

DIA’s AI-based analysis is currently being used in some 20 markets ب including North America and Europe (China also says that a partner may use its software as part of its device). Approved) Marketing strategy involving working with channel partners (such as GE, Philips and Konica Manolta) who offer software as an addition to their ultrasound or PACS system. ۔

According to Goldman Aslan, at this stage some 3,000+ end users have access to its software.

“Our technology vendor is neutral and cross-platform so it runs on any ultrasound device or healthcare IT system. That’s why you can see that with both of our device companies as well as healthcare IT / PACS companies. There are over 10 partnerships. There is no other startup in this space that I know has these capabilities, commercial traction or many FDA / CE AI based solutions, “she adds.” To date, we have 7 FDA / CE approved solutions for cardiac and abdominal areas, and more are on the way. “

AI’s performance is certainly as good as the data set on which it has been trained. And utility in the healthcare space is a particularly important factor دیکھ given that any bias in training data can lead to a flawed model that misdiagnoses or underestimates the risk of disease in patient groups. / Estimates higher which were not well represented in the training statistics.

Asked how his AI was trained to find key details in ultrasound imaging, Goldman Aslan told TechCrunch: “We have access to hundreds of thousands of ultrasound images through a number of medical facilities. So we have the ability to move faster from one automated area to another. “

“We collect diverse population data with data from different pathologies as well as different devices,” he added.

“Garbage in the trash is a phrase. The key is not to bring in the trash,” he told us. “Our datasets have been tagged and rated by many doctors and technicians, each with many years of experience.

“We also have a strong rejection system that rejects images taken incorrectly. In this way we control the issue of data retrieval.

It is worth noting that the FDA approval obtained by the DIA is 510 (K) Class II approval اور and Goldman Aslan has confirmed to us that it has received FDA approval for its products. Apply (and do not intend) to apply for market approval (PMA). .

Route 510 (k) is widely used to gain approval for the insertion of a variety of medical devices into the US market. However, it has been criticized as a light-hearted government – and certainly not scrutinized at the same level as the stricter PMA process.

The broader point is that regulation of rapidly evolving AI technologies lags behind the developments in which they are being applied ب including those that are rapidly advancing in the healthcare space where there are certainly many Big promises, but serious risks if they fail to live. Shiny Marketing – This means that there is still a difference between the promises made by the device makers and the regulatory oversight of their tools.

In the EU, for example, the CE scheme – which sets certain health, safety and environmental standards for devices – may require a manufacturer to declare itself compatible, without any independent verification. I meet the standards they claim. Some medical devices may require a degree of independent assessment of compliance under the CE scheme. But it is not considered a strict government to manage the security of modern technologies like AI.

Therefore, the EU is now working on introducing an additional layer of synergies for AI applications, especially those considered ‘high risk’ under the forthcoming Artificial Intelligence Act.

Matters of healthcare use, such as DIA’s AI-based ultrasound analysis, will certainly fall under this category, so there will be some additional regulatory requirements under the AIA. For now, though, the proposal on the table is being debated by co-EU lawmakers, and a dedicated regulatory system for high-risk AI applications is still in place in the region.

Leave a Reply

Your email address will not be published.