LEVERAGING AI & HYPERSPECTRAL IMAGING TO PROVIDE EARLY INSIGHTS INTO SKIN CANCER
Skin cancer is responsible for 1 death per hour which accounts for more than 50% of all cancers combined. The most aggressive and hazardous skin cancer type is Melanoma, which represents 4% of all skin cancer lesions. Melanoma diagnosis rates are continuously increasing however the number of Dermatologists is decreasing relative to the population. This means that Dermatologists do not have the capacity to regularly examine the general population using conventional techniques. At present, multiple repetitive examinations are required in order to safely determine skin abnormalities. Most systems that are currently available to doctors are optical or digital dermoscopes, operating in the visible spectrum. Such devices have limited diagnostic and predictive accuracy.
SpectroX project exploited the fact that Hyperspectral imaging is able to reveal rich information, both in and under the surface of skin, well beyond the capabilities of the human eye. With the additional aid of Artificial Intelligence (AI) and deep learning technology, this has enabled the multi-disciplinary team of Computer Scientists and dermatologists at Althexis to develop this breakthrough technology that can detect and classify skin conditions before they become they even become visible thus enabling earlier intervention resulting in improved patient outcomes.
SpectroX has proven to be highly accurate, easy to use as an early diagnostic system that will revolutionize the way in which skin conditions are detected. SpectroX significantly reduces the examination time by half compared to current solutions, while making a previously inaccessible technology, affordable without any compromises.
This is the key to scale population screening, which will lead to saving lives and medical costs.
SpectroX is a full solution for the dermatologist from the handheld image capture device with an AI enabled software capable of capturing and making skin image classifications for various skin conditions. It also includes full patient records, electronic GDPR consent forms, patient metadata. visits and images taken over time of skin lesions and moles. All data is uploaded to the Althexis Secure Cloud, enabling AI models to be continuously updated and distributed automatically to all the systems in the clinics. This process enables every dermatologist and their patients to gain from the latest collective insights and expertise and enables single-shot predictions of unprecedented accuracy, compared to the conventional procedures.
AI has rapidly evolved and matured and is used in a growing number of healthcare solutions providing outstanding performance. Spectral Imaging cameras enable precise and consistent image acquisitions, as the capture procedure is properly calibrated and represents an actual physical measurement. Althexis combines these along with its secure Cloud technologies to bring about the next generation dermoscope, with a special emphasis on the early detection of melanoma.
FED4SAE project has been critical in accelerating the efforts of the company to develop and deploy working devices, based on CSEM’s Hyperspectral Technologies that captured rich hyperspectral images from patients which has enabled the AI systems to be developed. This has enabled Althesis to validate their business model and reach the market faster. The collaborative network within FED4SAE provided the key enabling technologies from CSEM and Intel also valuable business support from BluMorpho.
SpectroX is the next generation dermoscope solution and it has been developed, deployed and validated within the FED4SAE Application Experiment. It is enabling Dermotologists to gain early detection of melanoma or certain skin conditions and that can significantly increase survival rates. SpectroX will also have a significant impact on the attitude towards skin screening from the general public and allow for a larger portion of the population to be tested. Currently (Q4’2020) Althexis is currently seeking Venture Capital funding, in order to scale its business by approaching new clients and develop new features within SpectroX.