Personalized Analytics is becoming essential in healthcare, stemming from the movement from fee-for-service to a value-based market. The need to preempt and prevent disease on a more personal level, rather than merely reacting to symptoms, has created a significant opportunity for machine learning-based applications. This "analytics of one" approach (using advanced mathematical models and artificial intelligence techniques) is already impacting several key areas: Prime examples include cardiac imaging analysis that aides physicians in assessing conditions, including heart attacks and coronary artery disease, and retinal image analysis to detect diabetic retinopathy. The anticipated goal for AI in healthcare is to enhance and expand the "four Ps" of care delivery – predictive, preventative, personalized and participatory. Predictive: Predictions have existed in healthcare for some decades now, as statistical models based on structured data sources.
A pair of smartglasses that automatically focus on objects which could be used by people with a variety of vision defects have been developed by researchers. Called'Autofocals', the glasses use depth-sensing cameras and eye-tracking technology which promise to keep objects in hyper-sharp focus at all times. Presbyopia is a common form of age-induced far-sightedness, where the lenses in the eyes become stiff and have trouble focusing on close-up objects. The condition typically kicks in at around age 45 and it affects more than a billion people and a key factor as to why many need to wear glasses in middle age. But a Stanford University team have developed a way to treat defects like this, so that when worn they mimic the natural'autofocus' mechanism of a healthy eye.
Machine learning AI has recently been used to distinguish between patients who are fit for corneal refractive surgery and those who are likely to experience post-operative complications. The referral for this procedure often goes misdiagnosed, but by using AI, these researchers have potentially created an accurate screening tool for the surgery. Their work was published on June 20 in the journal npj Digital Medicine. Refractive surgery, such as LASIK, utilize lasers to reshape the cornea in treating conditions such as near and farsightedness, and astigmatism. It is essential to screen candidates for these operations to prevent adverse outcomes, but there are no existing screening methods that address the possibility of improper diagnosis.
Eyes are more than the "windows to the soul." As such, ocular health and neurological health are intertwined. The most skilled ophthalmologists can read ocular scans to not only look for eye disease, but also traces of a host of neurological disorders. Voxeleron is using artificial intelligence and machine learning to, as they put it, "democratize expertise." Their algorithms hold the promise of delivering expert-level diagnostic capabilities to any lab with a scanning device.
The joint effort comes as health systems are stepping up adoption and investment in data analytics, including predictive analytics and AI. In recent survey of CIOs, CTOs and chief analytics officers conducted by the Deloitte Center for Health Solutions, 84% said such technology will be extremely important to their organization's strategy over the next three years. Other healthcare sectors are investing in AI as well, giving rise to potential safety, efficacy and ethical issues as the technology is more frequently used. One year ago, FDA approved the first autonomous AI diagnostic system for sale in the U.S. The cloud-based IDx-DR software detects diabetic retinopathy in images taken by retinal cameras. And in February, Verily, the life sciences arm of Google parent Alphabet, launched an eye disease screening algorithm at Aravind Eye Hospital in Madurai, India.
A 3D bio printer is being used to create human corneas at Florida A&M University. Research led by Florida A&M University Pharmaceutics Professor Mandip Sachdeva has resulted in the creation of the first high throughput printing of human cells in a 3D print of the cornea in the U.S. The scientific breakthrough -- created in two research laboratories in the Dyson Pharmacy Building on campus -- could lead to far-reaching advancements in the medical field, from transplants to testing of new cornea-relief products to cornea wound treatment. Sachdeva, along with Shallu Kutlehria, a graduate assistant in the College of Pharmacy and Pharmaceutical Sciences, and research assistant Paul Dinh, are completing a white paper to be submitted later this month for journal publication. The cornea advancement is an outgrowth of research Sachdeva has been engaged in with the help of a grant in 2017 from the National Science Foundation to Florida A&M and the FAMU-FSU College of Engineering. A link has been posted to your Facebook feed.
On March 26, at the Francis Crick Institute research centre in London, WIRED Editor Greg Williams kicked off WIRED Health 2019. The event, in its sixth edition, features talks, interviews and panel discussions with the world's most inspiring healthcare innovators, entrepreneurs, and researchers, in order to shed light on what lies ahead for the medicine and healthcare sectors. "We wanted to bring together some amazing innovators in the worlds of healthcare, pharma and technology to look at what the future looks like for health in the age of artificial intelligence, virtual reality, big data and genomics," Williams said in his opening remarks on the main stage. Over the following hours, the speakers took the floor, bringing to life an impressive collection of eye-opening stories about what it takes to make innovation happen. "Discovery research is going to create the future," said Paul Nurse, director of the Francis Crick Institute.
Chart-based visual acuity measurements are used by billions of people to diagnose and guide treatment of vision impairment. However, the ubiquitous eye exam has no mechanism for reasoning about uncertainty and as such, suffers from a well-documented reproducibility problem. In this paper we uncover a new parametric probabilistic model of visual acuity response based on measurements of patients with eye disease. We present a state of the art eye exam which (1) reduces acuity exam error by 75\% without increasing exam length, (2) knows how confident it should be, (3) can trace predictions over time and incorporate prior beliefs and (4) provides insight for educational Item Response Theory. For patients with more serious eye disease, the novel ability to finely measure acuity from home could be a crucial part in early diagnosis. We provide a web implementation of our algorithm for anyone in the world to use.
Diagnosis and treatment guidance are aided by detecting relevant biomarkers in medical images. Although supervised deep learning can perform accurate segmentation of pathological areas, it is limited by requiring a-priori definitions of these regions, large-scale annotations, and a representative patient cohort in the training set. In contrast, anomaly detection is not limited to specific definitions of pathologies and allows for training on healthy samples without annotation. Anomalous regions can then serve as candidates for biomarker discovery. Knowledge about normal anatomical structure brings implicit information for detecting anomalies. We propose to take advantage of this property using bayesian deep learning, based on the assumption that epistemic uncertainties will correlate with anatomical deviations from a normal training set. A Bayesian U-Net is trained on a well-defined healthy environment using weak labels of healthy anatomy produced by existing methods. At test time, we capture epistemic uncertainty estimates of our model using Monte Carlo dropout. A novel post-processing technique is then applied to exploit these estimates and transfer their layered appearance to smooth blob-shaped segmentations of the anomalies. We experimentally validated this approach in retinal optical coherence tomography (OCT) images, using weak labels of retinal layers. Our method achieved a Dice index of 0.789 in an independent anomaly test set of age-related macular degeneration (AMD) cases. The resulting segmentations allowed very high accuracy for separating healthy and diseased cases with late wet AMD, dry geographic atrophy (GA), diabetic macular edema (DME) and retinal vein occlusion (RVO). Finally, we qualitatively observed that our approach can also detect other deviations in normal scans such as cut edge artifacts.
The purpose of this retrospective study is to measure machine learning models' ability to predict glaucoma drainage device failure based on demographic information and preoperative measurements. The medical records of sixty-two patients were used. Potential predictors included the patient's race, age, sex, preoperative intraocular pressure, preoperative visual acuity, number of intraocular pressure-lowering medications, and number and type of previous ophthalmic surgeries. Failure was defined as final intraocular pressure greater than 18 mm Hg, reduction in intraocular pressure less than 20% from baseline, or need for reoperation unrelated to normal implant maintenance. Five classifiers were compared: logistic regression, artificial neural network, random forest, decision tree, and support vector machine.