Till some time back, AI in medicine and medical imaging was a distant thought. Today, AI is filtering into almost every aspect of medicine, bringing forth many approaches by which humans can interact with data and images in a seamless and automated manner. As the market blooms, sorting data and performing actual diagnostics and detection has been accelerated by startup companies and industry giants leveraging AI in their imaging solutions. Already working in this area is Cardiff-based MedaPhor—a firm providing intelligent ultrasound software platforms to unlock the potential of ultrasound for all medical practitioners.
CTO of the firm, Nicholas Sleep says, “Today, one of the major challenges facing physicians and clinicians is simplifying ultrasound so more medical professionals can use what is the fastest, safest and cheapest diagnostic tool in medicine”. To this end, MedaPhor has developed ScanNav, an AI system that links to standard ultrasound machines. Sleep extols, “We believe ScanNav is the world’s first AI system to support scanning in obstetric ultrasound by helping sonographers capture high-quality images that meet clinical requirements.”
The first product —ScanNav Audit— supports clinicians scanning under the UK Fetal Anomaly Scanning Programme (FASP) protocol. Here, images recorded during a 20-week obstetric scan are verified in real-time as the sonographer or doctor scan the patient. With 100 percent of images captured, ScanNav Audit aims to raise the standard of image acquisition and subsequent diagnosis by clinicians. This software is CE marked and already being used in two leading UK hospitals. The second product in the range is called ScanNav Autocapture which is designed to use AI to improve sonographer efficiency and to enable them to scan more patients during a session. The product automatically evaluates the live imaging data being acquired by the ultrasound machine and captures images that adhere to the agreed protocol criteria. Sleep expects this to reduce the required scanning time for complicated protocol-based scans dramatically.
Additionally, MedaPhor is working on the development of the next generation ScanNav platform- ScanNav Assist aims to confirm that key anatomical regions have been fully scanned, as well as highlighting any abnormalities on the ultrasound image for further investigation by the clinician.
ScanNav is the world’s first AI system to support scanning in obstetric ultrasound by helping sonographers capture high-quality images that meet clinical requirements
Medaphor is establishing partnerships with many peers of the industry to continually improve its AI image analysis platform. To this end, Sleep said that AI companies need access to large image libraries to teach their algorithms from as wide a range of anatomies as possible. “In this regard, we are trying to establish relationships with leading clinical institutions to allow us to develop the next generation of cost-effective and agile medical tools.”
The key to any AI software disrupting the market is its performance in the clinic. “With ScanNav we listened to the feedback we received from sonographers, and by adapting our technology to enhance their existing workflow, they now feel more confident in their scanning and see ScanNav as a supportive tool, rather than a threat.”
MedaPhor is also working on several other proprietary products for interventional ultrasound (helping make it easier to use ultrasound regional anesthesia) as well as its ultimate goal of enabling anyone to scan themselves at home with fully automated devices. “It’s an exciting time for the company, and we definitely feel we are leading the way in bringing AI software to market that can be truly disruptive to the field of ultrasound scanning,” states Sleep.