Biosignals, Robotics, and Rehabilitation – IEEE Spectrum

This sponsored article is dropped at you by NYU Tandon School of Engineering.

To handle immediately’s well being challenges, particularly in our growing older society, we should develop into extra clever in our approaches. Clinicians now have entry to a variety of superior applied sciences designed to help early prognosis, consider prognosis, and improve affected person well being outcomes, together with telemedicine, medical robots, powered prosthetics, exoskeletons, and AI-powered sensible wearables. However, many of those applied sciences are nonetheless of their infancy.

The perception that advancing expertise can enhance human well being is central to analysis associated to medical system applied sciences. This kinds the center of analysis for Prof. S. Farokh Atashzar who directs the Medical Robotics and Interactive Intelligent Technologies (MERIIT) Lab on the NYU Tandon School of Engineering.

Atashzar is an Assistant Professor of Electrical and Computer Engineering and Mechanical and Aerospace Engineering at NYU Tandon. He can be a member of NYU WIRELESS, a consortium of researchers devoted to the subsequent era of wi-fi expertise, in addition to the Center for Urban Science and Progress (CUSP), a middle of researchers devoted to all issues associated to the way forward for fashionable city life.

Atashzar’s work is devoted to creating clever, interactive robotic, and AI-pushed assistive machines that may increase human sensorimotor capabilities and enable our healthcare system to transcend pure competences and overcome physiological and pathological boundaries.

Stroke detection and rehabilitation

Stroke is the main explanation for age-associated motor disabilities and is turning into extra prevalent in youthful populations as nicely. But whereas there’s a burgeoning market for rehabilitation units that declare to speed up restoration, together with robotic rehabilitation methods, suggestions for a way and when to make use of them are based mostly totally on subjective analysis of the sensorimotor capacities of sufferers in want.

Atashzar is working in collaboration withJohn-Ross Rizzo, affiliate professor of Biomedical Engineering at NYU Tandon and Ilse Melamid Associate Professor of rehabilitation medication on the NYU School of Medicine and Dr. Ramin Bighamian from the U.S. Food and Drug Administration to design a regulatory science software (RST) based mostly on knowledge from biomarkers with a view to enhance the evaluation processes for such units and how greatest to make use of them. The crew is designing and validating a sturdy restoration biomarker enabling a primary-ever stroke rehabilitation RST based mostly on exchanges between areas of the central and peripheral nervous methods.

S. Farokh Atashzar is an Assistant Professor of Electrical and Computer Engineering and Mechanical and Aerospace Engineering at New York University Tandon School of Engineering. He can be a member of NYU WIRELESS, a consortium of researchers devoted to the subsequent era of wi-fi expertise, in addition to the Center for Urban Science and Progress (CUSP), a middle of researchers devoted to all issues associated to the way forward for fashionable city life, and directs the MERIIT Lab at NYU Tandon.NYU Tandon

In addition, Atashzar is collaborating with Smita Rao, PT, the inaugural Robert S. Salant Endowed Associate Professor of Physical Therapy. Together, they intention to establish AI-pushed computational biomarkers for motor management and musculoskeletal injury and to decode the hidden advanced synergistic patterns of degraded muscle activation utilizing knowledge collected from floor electromyography (sEMG) and excessive-density sEMG. In the previous few years, this collaborative effort has been exploring the fascinating world of “Nonlinear Functional Muscle Networks” — a brand new computational window (rooted in Shannon’s data idea) into human motor management and mobility. This synergistic community orchestrates the “music of mobility,” harmonizing the synchrony between muscle groups to facilitate fluid motion.

But rehabilitation is barely one of many analysis thrusts at MERIIT lab. If you may stop strokes from taking place or reoccurring, you may head off the issue earlier than it occurs. For Atashzar, an enormous clue might be the place you least count on it: in your retina.

Atashzar together with NYU Abu Dhabi Assistant Professor Farah Shamout, are engaged on a mission they name “EyeScore,” an AI-powered expertise that makes use of non-invasive scans of the retina to foretell the recurrence of stroke in sufferers. They use optical coherence tomography — a scan of the again of the retina — and observe adjustments over time utilizing superior deep studying fashions. The retina, hooked up on to the mind via the optic nerve, can be utilized as a physiological window for adjustments within the mind itself.

Atashzar and Shamout are at the moment formulating their hybrid AI mannequin, pinpointing the precise adjustments that may predict a stroke and recurrence of strokes. The end result will have the ability to analyze these pictures and flag probably troublesome developments. And because the scans are already in use in optometrist workplaces, this life-saving expertise might be within the arms of medical professionals prior to anticipated.

Preventing downturns

Atashzar is using AI algorithms for makes use of past stroke. Like many researchers, his gaze was drawn to the biggest medical occasion in latest historical past: COVID-19. In the throes of the COVID-19 pandemic, the very bedrock of world healthcare supply was shaken. COVID-19 sufferers, prone to swift and extreme deterioration, introduced a major problem for caregivers.

Especially within the pandemic’s early days, when our grasp of the virus was tenuous at greatest, predicting affected person outcomes posed a formidable problem. The merest tweaks in admission protocols held the facility to dramatically shift affected person fates, underscoring the necessity for vigilant monitoring. As healthcare methods groaned below the pandemic’s weight and contagion fears loomed, outpatient and nursing heart residents have been steered towards distant symptom monitoring through telemedicine. This cautious strategy sought to spare them pointless hospital publicity, permitting in-individual visits just for these within the throes of grave signs.

But whereas a lot of the pandemic’s analysis highlight fell on diagnosing COVID-19, this research took a unique avenue: predicting affected person deterioration sooner or later. Existing research usually juggled an array of information inputs, from advanced imaging to lab outcomes, however did not harness knowledge’s temporal elements. Enter this analysis, which prioritized simplicity and scalability, leaning on knowledge simply gathered not solely inside medical partitions but in addition within the consolation of sufferers’ houses with using easy wearables.

S. Farokh Atashzar and colleagues at NYU Tandon are utilizing deep neural community fashions to evaluate COVID knowledge and attempt to predict affected person deterioration sooner or later.

Atashzar, alongside together with his Co-PI of the mission Yao Wang, Professor of Biomedical Engineering and Electrical and Computer Engineering at NYU Tandon, used a novel deep neural community mannequin to evaluate COVID knowledge, leveraging time sequence knowledge on simply three very important indicators to foresee COVID-19 affected person deterioration for some 37,000 sufferers. The final prize? A streamlined predictive mannequin able to aiding scientific determination-making for a large spectrum of sufferers. Oxygen ranges, heartbeats, and temperatures shaped the trio of significant indicators below scrutiny, a selection propelled by the ubiquity of wearable tech like smartwatches. A calculated exclusion of sure indicators, like blood strain, adopted, as a result of their incompatibility with these wearables.

The researchers utilized actual-world knowledge from NYU Langone Health’s archives spanning January 2020 to September 2022 lent authenticity. Predicting deterioration inside timeframes of three to 24 hours, the mannequin analyzed very important signal knowledge from the previous 24 hours. This crystal ball aimed to forecast outcomes starting from in-hospital mortality to intensive care unit admissions or intubations.

“In a situation where a hospital is overloaded, getting a CT scan for every single patient would be very difficult or impossible, especially in remote areas when the healthcare system is overstretched,” says Atashzar. “So we are minimizing the need for data, while at the same time, maximizing the accuracy for prediction. And that can help with creating better healthcare access in remote areas and in areas with limited healthcare.”

In addition to addressing the pandemic on the micro stage (people), Atashzar and his crew are additionally engaged on algorithmic options that may help the healthcare system on the meso and macro stage. In one other effort associated to COVID-19, Atashzar and his crew are creating novel probabilistic fashions that may higher predict the unfold of illness when taking into consideration the consequences of vaccination and mutation of the virus. Their efforts transcend the basic small-scale fashions that have been beforehand used for small epidemics. They are engaged on these massive-scale advanced fashions with a view to assist governments higher put together for pandemics and mitigate speedy illness unfold. Atashzar is drawing inspiration from his lively work with management algorithms utilized in advanced networks of robotic methods. His crew is now using comparable strategies to develop new algorithmic instruments for controlling unfold within the networked dynamic fashions of human society.

A state-of-the-artwork human-machine interface module with wearable controller is one among many multi-modal applied sciences examined in S. Farokh Atashzar’s MERIIT Lab at NYU Tandon.NYU Tandon

Where minds meet machines

These tasks signify solely a fraction of Atashzar’s work. In the MERIIT lab, he and his college students construct cyber-bodily methods that increase the performance of the subsequent-era medical robotic methods. They delve into haptics and robotics for a variety of medical purposes. Examples embrace telesurgery and telerobotic rehabilitation, that are constructed upon the capabilities of subsequent-era telecommunications. The crew is particularly within the software of 5G-based mostly tactile web in medical robotics.

Recently, he acquired a donation from the Intuitive Foundation: a Da Vinci analysis equipment. This state-of-the-artwork surgical system will enable his crew to discover methods for a surgeon in a single location to function on a affected person in one other—whether or not they’re in a unique metropolis, area, and even continent. While a number of researchers have investigated this imaginative and prescient previously decade, Atashzar is particularly concentrating on connecting the facility of the surgeon’s thoughts with the autonomy of surgical robots – selling discussions on methods to share the surgical autonomy between the intelligence of machines and the thoughts of surgeons. This strategy goals to cut back psychological fatigue and cognitive load on surgeons whereas reintroducing the sense of haptics misplaced in conventional surgical robotic methods.

Atashzar poses with NYU Tandon’s Da Vinci analysis equipment. This state-of-the-artwork surgical system will enable his crew to discover methods for a surgeon in a single location to function on a affected person in one other—whether or not they’re in a unique metropolis, area, and even continent.NYU Tandon

In a associated line of analysis, the MERIIT lab can be specializing in reducing-edge human-machine interface applied sciences that allow neuro-to-system capabilities. These applied sciences have direct purposes in exoskeletal units, subsequent-era prosthetics, rehabilitation robots, and presumably the upcoming wave of augmented actuality methods in our sensible and linked society. One widespread important problem of such methods which is concentrated by the crew is predicting the supposed actions of the human customers via processing alerts generated by useful conduct of motor neurons.

By fixing this problem utilizing superior AI modules in actual-time, the crew can decode a person’s motor intentions and predict the supposed gestures for controlling robots and digital actuality methods in an agile and sturdy method. Some sensible challenges embrace making certain the generalizability, scalability, and robustness of those AI-pushed options, given the variability of human neurophysiology and heavy reliance of basic fashions on knowledge. Powered by such predictive fashions, the crew is advancing the advanced management of human-centric machines and robots. They are additionally crafting algorithms that keep in mind human physiology and biomechanics. This requires conducting transdisciplinary options bridging AI and nonlinear management theories.

Atashzar’s work dovetails completely with the work of different researchers at NYU Tandon, which prizes interdisciplinary work with out the silos of conventional departments.

“Dr. Atashzar shines brightly in the realm of haptics for telerobotic medical procedures, positioning him as a rising star in his research community,” says Katsuo Kurabayashi, the brand new chair of the Mechanical and Aerospace Engineering division at NYU Tandon. “His pioneering research carries the exciting potential to revolutionize rehabilitation therapy, facilitate the diagnosis of neuromuscular diseases, and elevate the field of surgery. This holds the key to ushering in a new era of sophisticated remote human-machine interactions and leveraging machine learning-driven sensor signal interpretations.”

This dedication to human well being, via the embrace of latest advances in biosignals, robotics, and rehabilitation, is on the coronary heart of Atashzar’s enduring work, and his unconventional approaches to age-outdated drawback make him an ideal instance of the strategy to engineering embraced at NYU Tandon.