In a major development for knee rehabilitation, researchers at Indian Institute of Technology (IIT) Ropar have introduced a revolutionary, low-cost solution to make Continuous Passive Motion (CPM) therapy more accessible to patients. The team’s newly patented innovation, the Completely Mechanical Passive Motion Machine for Knee Rehabilitation, is set to transform post-surgical recovery, especially in resource-limited areas.

Unlike traditional motorized CPM devices, which are expensive and reliant on electricity, the new machine operates entirely through mechanical means. Utilizing a piston and pulley system that stores air as the user pulls a handle, the device enables smooth, controlled knee motion to aid in rehabilitation. This design eliminates the need for electricity, batteries, or motors, making the machine lightweight, portable, and environmentally friendly.

The mechanical CPM machine addresses a key barrier to knee therapy: the high cost and power dependence of conventional electric machines. It offers a viable alternative for patients, particularly in rural and off-grid areas, where access to electricity is often unreliable. Its portability also enables patients to continue their therapy at home, reducing the need for frequent hospital visits or prolonged stays.

Knee rehabilitation is crucial for patients recovering from surgeries, as CPM therapy helps improve joint mobility, reduce stiffness, and speed up recovery. With this new device, IIT Ropar’s researchers are offering a cost-effective, sustainable option that could improve the lives of countless patients, especially in India, where advanced medical technology can be scarce in rural regions.

Lead researcher Dr. Abhishek Tiwari, along with his team members Suraj Bhan Mundotiya and Dr. Samir C. Roy, expressed optimism about the machine’s potential. “This device has the power to revolutionize knee rehabilitation, particularly in areas where access to sophisticated medical equipment is limited. It’s designed to be an affordable and eco-friendly solution that not only aids in recovery but also minimizes environmental impact,” said Dr. Tiwari.

The introduction of this innovative mechanical CPM machine marks a significant step toward democratizing healthcare and improving rehabilitation outcomes globally.

Researchers at the University of Cambridge have developed a machine learning algorithm capable of accurately detecting heart murmurs in dogs—a critical indicator of cardiac disease, particularly prevalent in smaller breeds like the King Charles Spaniel. This innovative approach has the potential to transform veterinary care, offering an accessible tool for early diagnosis and treatment of heart conditions.

Heart murmurs are a key sign of mitral valve disease, the most common heart issue affecting adult dogs. Statistically, approximately one in every 30 dogs seen by a veterinarian presents with a heart murmur, with higher rates observed in small breeds and older dogs. Given the frequency of such conditions, timely detection is essential. Early intervention can significantly enhance a dog’s quality of life and longevity, making effective screening methods vital for veterinarians.

Dr. Andrew McDonald, the study’s first author from the Department of Engineering at Cambridge, emphasized the importance of early detection, according to a statement issued by the University: “Heart disease in humans is a huge health issue, but in dogs it’s an even bigger problem. Most smaller dog breeds will have heart disease when they get older, but obviously dogs can’t communicate in the same way that humans can, so it’s up to primary care vets to detect heart disease early enough so it can be treated.”

The Algorithm’s Development

The research team began with an algorithm initially designed for human heart sound analysis. Recognizing the similarities between mammalian heart function, they adapted this technology to analyze audio recordings from digital stethoscopes used on dogs. The algorithm demonstrated an impressive sensitivity of 90% in detecting heart murmurs, a level of accuracy comparable to that of expert cardiologists.

Professor Anurag Agarwal, the lead researcher and an expert in acoustics and bioengineering, noted the absence of a dedicated database for canine heart sounds. “As far as we’re aware, there are no existing databases of heart sounds in dogs, which is why we started out with a database of heart sounds in humans,” he explained in a statement issued by the University of Cambridge. “Mammalian hearts are fairly similar, and when things go wrong, they tend to go wrong in similar ways.”

The team refined the algorithm to not only detect but also grade heart murmurs

To build a robust dataset, the researchers collected heart sound data from nearly 800 dogs undergoing routine examinations at four veterinary specialist centers across the UK. Each dog received a thorough physical examination and an echocardiogram performed by a cardiologist, who graded any detected murmurs and identified underlying cardiac issues. This effort resulted in the largest dataset of dog heart sounds ever compiled.

Expanding the Dataset for Better Outcomes

Co-author Professor Jose Novo Matos, a small animal cardiology specialist, highlighted the need for diverse data to improve the algorithm’s effectiveness: “Mitral valve disease mainly affects smaller dogs, but to test and improve our algorithm, we wanted to get data from dogs of all shapes, sizes, and ages. The more data we have to train it, the more useful our algorithm will be, both for vets and for dog owners.”

The team refined the algorithm to not only detect but also grade heart murmurs, distinguishing between mild and advanced disease requiring further intervention. This innovation aims to empower general veterinarians, reducing the need for expensive specialized scans and consultations with cardiologists.

Promising Results and Future Implications

The algorithm’s performance was encouraging: it aligned with cardiologists’ assessments in over half of the cases, and in 90% of instances, it was within one grading unit of the cardiologist’s evaluation. Dr. McDonald pointed out the practical implications of these findings: “The grade of heart murmur is a useful differentiator for determining next steps and treatments, and we’ve automated that process.”

Novo Matos remarked on the transformative potential of this technology, seeing it as a supportive tool rather than a job threat. “So many people talk about AI as a threat to jobs, but for me, I see it as a tool that will make me a better cardiologist,” he said. With the veterinary profession facing time constraints and a shortage of specialists, this algorithm could streamline the process of identifying dogs that need urgent care.

A Path Forward for Veterinary Medicine

The researchers’ ultimate goal is to equip veterinarians with the means to make informed decisions regarding treatment, enhancing the quality of life for their canine patients. “Knowing when to medicate is so important, in order to give dogs the best quality of life possible for as long as possible,” said Agarwal.

Supported by organisations such as the Kennel Club Charitable Trust and the Medical Research Council, this research marks a significant step forward in the use of machine learning for veterinary applications. As technology continues to evolve, it holds the promise of not only advancing animal health but also improving the human-animal bond through better care and understanding.

In an enlightening conversation with EdPublica, Sajikumar Sreedharan, Associate Professor at the NUS Yong Loo Lin School of Medicine, Singapore, shares his research insights on memory formation and the transition from short-term to long-term memory. His areas of research include aging and neurodegeneration, the neural basis of long-term memory (LTM), and synaptic tagging and capture (STC) as an elementary mechanism for storing LTM in neural networks. He also explores metaplasticity as a compensatory mechanism for improving memory in neural networks. With a career spanning over two decades, Prof. Sreedharan discusses his key findings, innovative methodologies, and the significance of receiving the “Investigator” award from the International Association for the Study of Neurons and Brain Diseases. Join us as he reflects on his journey and the collaborative spirit that drives his research.

Edited Excerpts:

Your research has been recognized for significantly advancing our understanding of memory formation. Could you elaborate on your key findings related to the transition from short-term to long-term memory?

I have been working in the field of learning and memory since 2000. My first mentor in neuroscience was Prof. T. Ramakrishna, the founder and first head of the Life Sciences Department at the University of Calicut, Kerala, India. He was a great motivator, and we often had insightful discussions about learning and memory in the evenings. I had the chance to work with him for my master’s dissertation, which was my first real research experience. Prof. Ramakrishna encouraged me to expand my knowledge further, and he connected me with Dr. Shobi Valeri, a senior researcher in Delhi at the time.

Dr. Shobi soon left for Germany to pursue his Ph.D. and recommended me to DRDO (Defence Research and Development Organisation). Dr. Shobi is now a senior scientist at the National Institute of Nutrition in Hyderabad. I worked at DRDO for a year before moving to Magdeburg, Germany, where I began my Ph.D. under Prof. Juletta Frey. She is well-known in the field of learning and memory, particularly for her research on the cellular mechanisms involved in forming associative memory.

In Prof. Frey’s lab, I discovered how different pieces of information can link together to form long-term memories. This work later inspired the development of many computational models of memory. After completing my Ph.D., I did my postdoctoral studies with Prof. Martin Korte in Braunschweig. There, I discovered how activating neurons before learning could enhance memory formation in the future, a process known as metaplasticity—an exciting and emerging area of neuroscience.

“Using animal models, we have uncovered the role of specific brain regions, like CA2 and CA1, in forming social and spatial memories—both of which are significantly affected by aging, neurodegenerative diseases, and mental health conditions

---

Since 2012, I have been working at the National University of Singapore, where I have focused more on aging, neurodegeneration, and mental health. Using animal models, we have uncovered the role of specific brain regions, like CA2 and CA1, in forming social and spatial memories—both of which are significantly affected by aging, neurodegenerative diseases, and mental health conditions.

How do you approach the study of molecular mechanisms in memory, and what methodologies do you find most effective?

In my lab, we approach research questions by examining them from different angles—molecular, cellular, behavioural, and system-level. We choose the most appropriate method depending on the specific question we’re investigating. I can’t say that one method is better than the others because each plays an important role in confirming our findings.

Recently, we’ve been using optogenetic and chemogenetic tools, which allow us to target and stimulate specific neurons. These methods are particularly helpful because they ensure precision in how we activate or deactivate brain cells.

Congrats on receiving the “Investigator” award from the International Association for the Study of Neurons and Brain Diseases. What does this recognition mean to you personally and professionally?

Thank you for your kind words. As a researcher, I feel proud and happy that my work is being recognized internationally. Professionally, this recognition is a significant motivation to continue pursuing my research.

This achievement is not just mine alone—I owe it to all my Ph.D. students, postdocs, and research technicians who have worked with me over the past 20 years. This award is for them as well.

How do you feel your work contributes to the broader scientific community, especially concerning memory impairments related to aging and mental health?

I am the Research Director of the Healthy Longevity Translational Research Programme at the School of Medicine, National University of Singapore, where we have more than 36 scientists working on various aspects of healthy aging. One of our key areas is brain health. Living a long life is not meaningful without a healthy brain.

I am one of the principal investigators studying how neural networks are impaired during aging and neurodegeneration. My wife, Dr. Sheeja Navakkode, is also a neuroscientist, focusing on Alzheimer’s disease using animal models. Neural networks undergo tremendous changes during aging and in various mental health conditions. Our goal is to correct or rewire neural network activity so that memory can be preserved with minimal damage, especially during conditions such as aging, Alzheimer’s Disease, and mental health disorders.

(Read the full interview in the upcoming December 2024 issue of EdPublica magazine.)