Biomedical imaging and artificial intelligence (3rd part)
Biomedical imaging and artificial intelligence (3rd part)
Applying Deep Learning to Biomedical Imaging
For the first and second part of this newsletter:
AI/ML solutions and tools for imaging
⚙ RetiSpec is a Toronto-based medical imaging company that uses AI and a retinal scan to help detect early signs of Alzheimer’s. Initially, their AI-based solution analyzes data-rich images captured with existing retinal imaging cameras available in most eye clinics. The data-rich images of the retina are captured using high resolution retinal imaging providing the basis for detailed analysis and more accurate results. Each pixel has over 100 colors instead of just 3 (i.e., red, green, blue) in a regular color image. Then automated Alzheimer’s detection uses AI to help clinicians detect Alzheimer’s, years before the emergence of clinical symptoms.
“Your eyes show the strength of your soul.” Paulo Coehlo
“Is the Eye the Window to Alzheimer’s?” Eliav Shaked, the CEO and co-founder of RetiSpec
In January 2024, the K2 Medical Research partnered with the Magruder Eye Institute and RetiSpec (2016) in a collaboration that aims to reduce time associated with recruiting eligible clinical trial candidates by using RetiSpec’s AI technology to detect neurological diseases through an eye exam.
⚙ HeartBeat.bio in Vienna, Austria is reinventing cardiac discovery with unique human organoid technologies, hardware engineering and AI. They have generated iPSC-derived, multi-cellular and lineage-controlled 3D cardiac tissue models from healthy donors or patients that recapitulate the physiology of the human heart in a way not seen before with conventional 2D or bioengineered model systems.
Founded in 2021, HeartBeat.bio secured in Pre-Series A financing €4.5 M funding for its cardioid drug discovery platform.
“Using AI to optimize for rapid neural imaging”
MIT CSAIL researchers are combining AI and electron microscopy to create a detailed brain network mapping, aiming to enhance connectomics research and clinical pathology.
⚙ Alpenglow Biosciences sheds new light on tissue analysis with an AI-powered 3D approach. With novel chemistry, a patented high-throughput light-sheet microscope and cloud-based AI analysis the company adds a new dimension to your pathology, away from the limitations of conventional 2D, slide-based analysis. Their research services include sample preparation, cloud-based data processing and storage and AI-powered 3D analysis, using its patented multi-resolution 3Di platform for 3D spatial biology.
Alpenglow Biosciences (US, 2018) has raised a total of $5.6M in funding over 3 rounds.
⚙ Auris Health engages in the development of robotics technology for medical applications with the MONARCH™ Platform and was acquired in 2019 by Ethicon, part of the Johnson & Johnson Medical Devices Companies. The Monarch Platform, is an endoscopic technology that promises to fight lung cancer by enabling physicians to diagnose hard-to-reach, small peripheral nodules earlier and with greater precision than ever before. The robotic system comprises a Monarch tower, cart, physical controller, and an intuitive physical and digital User Interface that combines traditional endoscopic views with computer-assisted navigation based on 3D models of the patient’s own lung anatomy. The Monarch Platform is an AI system that navigates with 3D models of the patient's lungs and then uploads each case to a cloud.
⚙ Coreline Soft, an AI software developer to analyze CT scans of the chest based in South Korea, has just received FDA and Korean Ministry of Food and Drug Safety (MFDS) clearance for its lung cancer detection software. The company claims that the software has achieved a sensitivity of 92.3% and a specificity of 80.4% in clinical trials.
In August 2023, Coreline Soft, announced it will sell its chest diagnosis solution Aview LCS PLUS—that provides information on the 3 main lung diseases: lung cancer, COPD and coronary artery calcification—in Switzerland through a strategic partnership with Swiss company HealthCare Connect (HCK).
For the complete list of FDA-cleared algorithms for medical devices:
“Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices”,
“Medical AI Evaluation” an aggregation of 141 FDA approved AI devices with reports on how each device was evaluated,
520 FDA-approved AI medical algorithms are now available in the US.
⚙ MedCognetics, a UTSW spinout, recently received FDA 510(k) approval for their breast cancer screening software: QmTRIAGE. Avenda Health, a UCLA spinout, recently scored FDA 510(k) approval for their prostate cancer care software: iQuest. Ultromics received FDA 510(k) clearance for a ML-based decision-support system used to help detect heart failure with preserved ejection fraction, a difficult diagnosis to make otherwise.
⚙ RetinAI (Bern, Switzerland, 2017) announced in 2020 a multi-year collaboration with Novartis under which RetinAI’s AI tools will be employed to support multiple projects in ophthalmology and digital health. RetinAI has a CE-marked web-based software platform, the RetinAI Discovery®, used to navigate imaging data, imaging biomarkers and track changes over time for some of the most common forms of eye diseases.
On December 2, 2022, Retin_AI announced the launch of Discovery CORE with AI for retinal fluid and layer segmentation. Discovery CORE and its AI models have been designed to accelerate data analysis and help clinical and academic researchers collaborate more efficiently in real time with their peers on medical and imaging datasets. The Discovery CORE uses RetinAI’s CE-marked fluid and layer segmentation AI models to quickly visualize and generate measures such as fluid volumes and retinal layer thickness on OCT volumes (Optical coherence tomography or OCT, is an imaging method used to generate a picture of the back of your eye, your retina).
In 2023, RetinAI announced a partnership with Boehringer Ingelheim in order to improve patient outcomes in geographic atrophy/GA—a progressive, advanced form of age-related macular degeneration and a leading cause of complete loss of sight, estimated to affect around 5 million people worldwide—by combining RetinAI’s Discovery platform and AI tools with Boehringer Ingelheim’s research in retinal diseases.
Retinai has raised a total of CHF 2.8M (€ 3M) in funding over 6 rounds.
⚙ Qynapse (France, 2015) is a medical technology company that unlocks the potential of quantitative imaging for diagnosis, prognosis and longitudinal monitoring of patients with Alzheimer's and other neurological diseases, allowing scientists to select patients, identify subgroups and conduct adaptive monitoring for clinical trials. They offer QyScore—for advanced diagnosis, drug efficacy and safety monitoring for central nervous system diseases—and QyPredict, an AI prediction technology for better targeted patient selection in clinical trials.
In 2022, Calyx—the eClinical and Regulatory solutions and services provider for solving complex data challenges in clinical research—announced it has entered into a partnership with Qynapse, to enable Calyx to adopt and deliver Qynapse's AI-powered neuroimaging analysis solutions, QyScore® and QyPredict® to its global pharmaceutical and biotech customers to develop new medical treatments for central nervous system disorders.
Qynapse has raised a total of $6.6M.
⚙ Blackford Analysis provides tools and services to unlock the value of imaging AI and improve patient outcomes for clinicians.
Blackford Platform™ is purpose built for integration with your existing systems while simplifying integrations and the management of multiple disparate AI applications and algorithms, which ultimately reduces the load on imaging platforms. New AI applications from their market-leading portfolio, as well as in-house developed AI applications, can easily be added to the image-processing platform, reducing implementation time, costs, and long-term maintenance. Long term performance is monitored 24/7 by Blackford Dashboard.
Blackford (Edinburgh, 2010) partnered with Nurea in 2023, a Bordeaux-based company building cardiovascular AI solutions. The two companies are engaged in a commercial partnership, in order to make Nurea’s PRAEVAorta® 2 solutions—that provides an automated reconstruction and analysis of the arterial tree from DICOM images—accessible to healthcare professionals through the Blackford platform.
Blackford Analysis has raised a total of $6.9M in funding over 6 rounds.