09 APRIL 2019
Nvidia and ACR collaborate to spur AI adoption in diagnostic radiology
Nvidia has partnered with the American College of Radiology (ACR) to offer an open artificial intelligence (AI) platform for the development of diagnostic algorithms.
Called Clara AI, the computational platform will be integrated with ACR Data Science Institute’s ACR AI-LAB. It will help radiologists devise and deploy medical imaging workflows and instruments, as well as create AI algorithms using their own data and facilities.
The programme was piloted for three months at Massachusetts General Hospital, Ohio State University (OSU), and the Center for Clinical Data Science (CCDS) at Brigham and Women’s Hospital.
Nvidia healthcare vice-president Kimberly Powell said: “Nvidia builds platforms that democratise the use of AI and we purpose-built the Clara AI toolkit to give every radiologist the opportunity to
“Our successful pilot with the ACR is the first of many that will make AI more accessible to the entire field of radiology.”
During the pilot study, OSU used the Clara AI toolkit to access a pre-trained model created by CCDS. The model labelled the university’s data for further testing and improving the algorithm.
This research led to the development of a cardiac computed tomography (CT) angiography model. The shared approach was said to have minimised algorithm training, validation and testing.
The software was also said to facilitate data aggregation, image annotation, image pre-processing and transformation, as well as algorithm transfer and computing.
OSU Wexner Medical Center radiology and medical imaging informatics department chair Richard White said: “Enabling a network of artificial intelligence between hospitals will create more robust algorithms, greater efficiencies and likely lead to better patient outcomes.”
Nvidia Clara platform also powers GE Healthcare’s Edison AI platform and the Nuance AI Marketplace, which support AI-LAB and help deploy AI within the radiology workflow.
08 APRIL 2019
Frost & Sullivan reveals predictions for patient monitoring device market
Business consultancy Frost & Sullivan has made a series of predictions for the future of the patient monitoring device market, following its recently published analysis of investments and trends in the industry.
The firm have suggested that artificial intelligence, brain-computer interfaces (BCI), wearables, smart prosthetics/implants, nano-robotics and smart fabrics are the top six trends in the industry. These predictions have been inspired by Frost & Sullivan’s recent analysis, Patient Monitoring Industry—Analysis of Investment and Trends, 2018.
As patient monitoring technology has improved, the ad hoc interval observations undertaken by nurses and other medical staff have in many cases been replaced by continuous monitoring devices that can track multiple biometrics. The six technologies outlined by Frost & Sullivan are all focused on providing real-time, measurable monitoring.
Recent developments in wearable technology include a wearable sensor to detect hidden anxiety and depression in children, developed in a joint research programme between the University of Vermont and the University of Michigan.
Meanwhile, researchers at the University of Chicago have been working on bringing sensory feedback to brain-controlled prosthetics, and at the University of Pennsylvania scientists have developed nano-robots so small that they can be injected into the body through a hypodermic needle.
A University of Massachusetts project has also been working on developing pyjamas made out of smart fabrics that monitor heartbeat, breathing and posture, and could be employed to monitor the health of cardiac patients.
Frost & Sullivan global director for advanced medical technologies Sowmya Rajagopalan said: “In the future, patient monitoring data will be combined with concurrent streams from numerous other sensors, as almost every life function will be monitored and its data captured and stored.
“The patient monitoring market is expected to be worth more than $350bn by 2025, as the focus is likely to move beyond device sales to solutions.”
The growing patient monitoring device market has caused a drastic surge in the amount of data available to clinicians, and many healthcare providers are turning to digital analytics to manage it all.
Predictive analytics not only present a comprehensive overview of a patient’s current health, but allows for the prediction of future illnesses. Such technologies attracted $566.3m in investments in 2018, according to Frost & Sullivan.
05 APRIL 2019
Australia to reform medical device regulatory system
The Australian Therapeutic Goods Administration (TGA) has announced plans to overhaul the regulatory system for medical devices in the country to increase patient safety.
An action plan aims to improve transparency and ensure manufacturers provide more information about their devices to patients and healthcare professionals.
The plan will improve the marketing authorisation process for new devices and the follow-up of existing devices.
Open public consultations will be conducted to obtain feedback on proposed policies, regulations and guidance materials.
Announcing the action plan, the TGA said: “It will fast track TGA’s implementation of medical device reforms already underway and consult on new ways to improve transparency and increase public and health professional confidence in the regulatory system.
“Involving consumers in the process is critical, noting that decisions on new policies or changes to the Therapeutic Goods Act will be required from Parliament prior to potential implementation.”
Under the new plan, regulations will be expanded to cover technologies such as 3D printed devices and software apps. The TGA will update the assessment process to address device cybersecurity risks and offer guidance on associated requirements for IT systems.
The regulator intends to identify problems with existing medical devices and enable quicker action. These systems will facilitate cross-checks with international regulators regarding any significant adverse events (AE) reports on specific products.
TGA will prioritise the traceability of devices across the healthcare supply chain. It plans to propose a device identifier system to track product history.
Following media reports about unsafe medical devices, TGA plans to team up with consumer groups to co-develop a strategy that would promote awareness about the regulatory system.
TGA said: “We will also strengthen consumer awareness of the responsibilities of the TGA, suppliers of medical devices and health professionals through a range of new consumer communication and education programmes.”
The regulator will publicly report review timeframes for new products to ensure timely access to consumers.
04 APRIL 2019
Researchers develop wearable biosensors that mimic skin’s properties
Researchers from universities in Binghamton and New York have developed a wearable biosensor.
This open-mesh electromechanical biosensor is designed to reflect the skin’s microarchitecture. It comprises a biological component and a physiochemical detector that tracks and analyses lactate and oxygen on the skin.
The sensor is also embedded with flexible gold sensor cables that mimic skin’s elasticity.
Binghamton University PhD student Matthew Brown said: “This topic was interesting to us because we were very interested in real-time, on-site evaluation of wound healing progress in the near future.
“Both lactate and oxygen are critical biomarkers to access wound-healing progression.”
The researchers hope that the sensor will integrate additional biomarkers in the future to increase its functionality. The sensors are expected to be incorporated into internal organs to assist in disease research.
Binghamton University biomedical engineering associate professor Ahyeon Koh said: “The bio-mimicry structured sensor platform allows free mass transfer between biological tissue and bio-interfaced electronics.
“Therefore, this intimately bio-integrated sensing system is capable of determining critical biochemical events while being invisible to the biological system or not evoking an inflammatory response.”
The researchers published their findings in the Biosensors and Bioelectronics journal. In the publication, the team noted that the new sensor was able to withstand common harsh, multiaxial stresses when assessed in systematic studies with variations in polymeric membranes.
03 APRIL 2019
FDA seeks feedback on regulatory framework changes for AI
The US Food and Drug Administration (FDA) has proposed a new regulatory framework for artificial intelligent medical devices.
To ensure that the machines only work for their approved purpose, FDA-cleared algorithms are currently locked to prevent the system from continuing to adapt and learn each time the code is used. The algorithm is instead trained manually.
Suggested framework changes will allow these continuously learning algorithms to be trained using real-world user data while also maintaining the safety and efficacy of the device.
FDA commissioner Scott Gottlieb said: “With AI, because the device evolves based on what it learns while it is in real-world use, we’re working to develop an appropriate framework that allows the software to evolve in ways to improve its performance while ensuring that changes meet our gold standard for safety and effectiveness throughout the product’s lifecycle-from premarket design throughout the device’s use on the market.”
The regulatory agency has published a discussion paper outlining the new framework that considers an algorithm’s performance, the manufacturer’s modification plans and the maker’s ability to manage and control modification risks.
The FDA intends to publish draft guidance based on the feedback received from the discussion paper.
AI algorithms are currently being used to screen diseases and offer treatment recommendations.
In April last year, the FDA cleared IDx’s AI-based medical device to detect diabetic retinopathy.
The agency also granted marketing authorisation to VizAI’s clinical decision support software in February last year.
02 APRIL 2019
Smith & Nephew to buy monitoring device maker Leaf Healthcare
Smith & Nephew has entered an agreement to acquire medical device company Leaf Healthcare, including its patient mobility monitoring device Leaf Patient Monitoring System.
Intended for pressure injury prevention, the system comprises a wireless, lightweight, wearable sensor that tracks an individual’s position, orientation, movement and activity. This allows remote monitoring of hospitalised patients.
Statistics reveal that approximately 2.5 million pressure injuries are treated per year at acute care facilities in the US, costing around $11bn.
Data captured by the system can be used to automate and document compliance with patients’ prescribed turn protocols for those who are at risk for pressure injuries.
An independent, randomised, controlled trial performed by Stanford University showed a 43% relative rise in turning protocol compliance in high-risk patients using the Leaf Patient Monitoring System.
The acquisition of this device is expected to complement Smith & Nephew’s Allevyn Life and Allevyn Gentle Border and Secura Skin Care products.
Smith & Nephew Advanced Wound Management president Simon Fraser said: “Consistent with our initial strategic investment, Smith & Nephew is focused on providing not just products to treat conditions, but also supporting clinicians with technologies designed for prevention, as well as treatment, and helping healthcare facilities reduce the cost of care.
“The Leaf Patient Monitoring System is highly complementary to Smith & Nephew’s existing wound portfolio and we are excited by the opportunities of expanding this product within our global portfolio.”
Over a period of two years, Smith & Nephew partnered with Leaf Healthcare as an exclusive distributor and strategic investor.
Leaf Healthcare co-founder and CEO Barrett Larson said: “We are excited by this new opportunity to deploy our award-winning pressure injury prevention technology through Smith & Nephew’s extensive advanced wound management portfolio.”
Last month, Smith & Nephew agreed to acquire regenerative medicine products developer Osiris Therapeutics for $660m. This deal is also intended to support the UK company’s wound management business.
01 APRIL 2019
Computational model to allow rare hereditary diseases diagnosis
Scientists at Lawson Health Research Institute in Canada have developed a new computational model that could assess epigenetic patterns to allow detection of rare hereditary diseases.
Prior research showed that the genome-wide assessment of DNA methylation could enable diagnosis of previously unknown neurodevelopmental or congenital conditions.
During the latest research, the scientists applied the new computational model to DNA samples obtained from 965 patients who had neurodevelopmental and congenital anomalies (ND/CAs) without a definitive diagnosis even after rigorous clinical genetic testing.
The new model is said to have helped in diagnosing several new cases among the subjects.
It identified 15 cases of syndromic Mendelian disorders, 12 with imprinting and trinucleotide repeat expansion disorders, and 106 with rare epi-variants.
Existing tests for patients with neurodevelopmental and congenital anomalies do not resolve numerous cases, noted the scientists.
Lawson Health Research Institute associate scientist Bekim Sadikovic said: “Many families spend years going through repeated testing and assessments in search of a diagnosis. This process is hard on patients and families, and is a great cost to our healthcare system.”
Limited treatments are also available for such disorders, but a specific diagnosis is expected to aid physicians in predicting the condition’s course, as well as improve planning and support for patients.
The research has been published in the American Journal of Human Genetics, where authors said that computational model facilitates concurrent detection of 14 syndromes using DNA methylation data with accuracy.
A statement from the publication read: “This study demonstrates that genomic DNA methylation analysis can facilitate the molecular diagnosis of unresolved clinical cases and highlights the potential value of epigenomic testing in the routine clinical assessment of ND/CAs.”
The scientists have licensed the new technology to Greenwood Genetics Laboratories in the US and Amsterdam University Medical Centre Laboratories in the European Union (EU).
29 MARCH 2019
Purdue University creates glutamate sensor to help monitor neurotrauma
Researchers at Purdue University have created a sensor that could track glutamate, which is associated with neurological conditions such as migraines, Alzheimer’s and Parkinson’s.
Currently being used as a research tool for animal testing, the sensor is being developed to monitor drug responses as part of treatments for neurotrauma.
Tracking glutamate levels has traditionally been considered difficult, hindering the development of new therapies. Previous devices have not been sensitive enough to detect the neurotransmitter, fast enough to track its increase or affordable enough for use in long-term projects.
Purdue University neuroscience and biomedical engineering professor Riyi Shi said: “When you feel like you’re running a fever, it doesn’t matter when you check your temperature, it will probably be the same for several hours.
“But a glutamate spike is so fast that if you don’t capture it at that moment, you miss the whole opportunity to get data.”
In-vivo measurement of glutamate levels would provide information on spinal cord injuries and the development of brain diseases.
Implanted on the spinal cord, the sensor is 3D-printed and laser-micromachined, enabling researchers to change its size, shape and orientation. This will enable it to carry out testing without the need for microfabrication.
Purdue University biomedical engineering assistant professor Hugh Lee said: “We wanted to create a low-cost and very fast way to build these sensors so that we can easily provide researchers with a means to measure glutamate levels in vivo.”
When tested in an animal model, the sensor was able to quickly capture spikes in glutamate. Current devices take 30 minutes.