Healthcare industry is one of the most happening industry in the world which is growing day by day. Scientists are developing various strategies and methodologies to fight against the diseases for the well being of the human mankind. The launching of Wearable Technology has opened up new windows for the development and advancements in the field of science and technology. Among these, development of wearable devices has captured great interest among physicists, chemists and material scientists alike.
Wearable Technology
Wearable technology, wearables, smart wear or skin electronics are smart electronic based devices (having micro-controllers) that are worn close to and/or on the skin surface, where they analyze, detect, and transmit information relating to body signals or ambient data and which in some cases allow immediate biofeedback to the wearer. In 1980s, the discovery of digital hearing aids put forward the beginning of digital methods and gave rise to a new category of medical devices. Today, development of wearable based sensors is considered as one of the hot research topic published by the analysis of American Chemical Society as per 2020 report. Nowadays, artificial intelligence and machine learning technologies based wearable devices are replacing the conventional devices in healthcare applications. In the recent years, it has been seen that AI based technology marked the beginning of an era relating to a more digitalised smart life.
Artificial Intelligence
A wide-ranging branch of computer science is concerned with building smart machines capable of performing tasks that typically require human intelligence. AI is an interdisciplinary science with multiple approaches, but advancements in machine learning and deep learning are creating a paradigm shift in virtually every sector of the tech industry.
Today, supported by sensors and artificial intelligence, wearable devices are being actively used for early diagnosis and monitoring chronic conditions. The use of smart wearables in the healthcare sector has been growing day by day. In overall, because of their mobility, accuracy, and ease of use, AI-powered wearables can be excellent aids for healthcare community. In the times of COVID-19, when providing proper healthcare facilities has become difficult, AI wearables are especially important.
Strong AI, is a machine’s ability to perform intellectual functioning and make data based decisions like a human being. Such type of intelligence is being used in big robotics systems, self driving transportation systems, etc. For healthcare wearables, we’re working with weak AI. It just enables machines to handle specific reasoning or problem-solving tasks and isn’t able to do human-like cognitive abilities. Visual perception, speech perception and even decision-making capabilities can be present in wearable AI applications, but none are at a high level. So, don’t expect AI-based wearables to perform full-fledged health monitoring or write a prescription based on your health conditions. These devices are unlikely to substitute doctors in the near future. But despite that, even now, the integration of wearable technologies and AI can greatly support doctors in their day-to-day activities and improve patients’ lives. Let’s have a look at various examples.
AI in Heart Monitoring
For about a decade, AI elements have already been used in cardiology. For example, AI was successfully used in cardiac imaging and diagnostic systems to analyze Electrocardiograph (ECG), work optimization, etc. Now, machine learning based algorithms found application in wearable heart monitors and smartphone apps, allowing laboratorians to monitor heart conditions remotely.
EKO
A California-based startup, ‘Eko’ which became famous for its digitalized stethoscopes, introduced an AI-powered platform that consists of advanced heart monitors and AI-powered reporting.
The platform helps identify early signs of heart disease and atrial fibrillation and shares data with the patient’s doctor. The AI algorithms analyze the patient’s heart condition through the heart beat and a single-lead ECG. Trained on multiple heart sound recording datasets, the AI algorithms use a deep neural connecting model for sound analysis. They are able to separate the cardiac beats and catch the tissue vibration caused by blood flow and identify murmur like sounds that are hard for the human ear to detect. Eko is aiming to expand its platform to more healthcare providers and launch an at-home monitoring program for cardiac patients.
AI in Predicting Woman’s Fertility
A health company named ‘Ava’ based in Zurich and San Francisco, introduced a solution that analyzes multiple health parameters to calculate the woman’s fertility peaks.
It consists of a mobile app, a smart sensor bracelet and a backend with machine learning and AI algorithms are kept together. They collect data on ovulation cycles, hormonal changes, body temperature, and other physiological parameters. The data is collected in a big database, processed, and analyzed using AI that delivers a report with insights and recommendations. This methodology has already been tested and validated in multiple clinical trials.
AI Wearables for the Blind and Visually Impaired
A couple of years ago, a US-based healthcare startup, Sunu, introduced a wrist band, a revolutionary healthcare wearable. The band was designed to help the blind or visually impaired to explore the environments and avoid collision with nearby obstacles.
With the help of sonar and echolocation, Sunu Band detects objects up to 16 feet or 5.5 meters away. Then haptic vibration feedback informs the user how close (or far away) they are from objects.
This band got popularized in the healthcare sector. But it doesn’t had AI features. Later with some features Google Glass came into the limelight. The device was supported to help hands-on workers with voice-activated assistance. Also, with the AI software toolkit from Envision, Google Glass 2 makes an excellent tool for blind and vision impaired people that helps them understand the environment. It uses camera to identify nearby (or far away) objects and AI software to recognize and describe them to a user.
Envision’s software is based on Optical Character Recognition (OCR) technology which is one of the fastest and most accurate one. With the app, the Glass can read text from posters, food packaging, barcodes, QR and display screens, and some handwritten text. This device can also recognize faces, identify object shapes, colors and describe surrounding scenes, and read complex documents. The application is multi-lingual (it supports around 60 languages) and has an excellent voice recognition user interface.
AI Tool For Deaf People
Starkey Hearing Technologies, a Minnesota based healthcare company, introduced Livio Edge AI, a multifunctional hearing aid equipped with natural language processing capabilities and gesture-recognition.
As per the reports, audibly impaired people usually suffer from losing balance. This could potentially put them at higher risk of falling, which can cause serious injury. The hearing aid uses also 3D motion and gesture detection sensors to capture the surrounding environment. Next, the data gets collected, analyzed with the AI-based app, which makes smart, immediate adjustments to the sound. This approach allows hearing-impaired people to catch a full perception of the surrounding world.
Source – Starkey
Detection of a Conversation’s Tone using Wearable AI System
Researchers from MIT’s Institute of Medical Engineering and Science (IMES) and Computer Science and Artificial Intelligence Laboratory (CSAIL) said that they’ve come closer to a solution: an AI based wearable system that can predict if a person’s mood is happy, sad, or neutral based on tones and speech patterns. They recorded 31 different conversations of several time intervals each, the team trained two algorithms on the data: One classified as the overall nature of a conversation either happy or sad, while the second classified each five-second block of every conversation as positive, negative, neutral, etc. “The system picks up on how, for example, the sentiment in the text transcription was more abstract than the raw accelerometer data,” says Alhanai. “It’s quite remarkable that a machine could approximate how we humans perceive these interactions, without significant input from us as researchers.” he says.
Our next step is to improve the algorithm’s emotional granularity so that it is more accurate at calling out boring, tense, and excited moments, rather than just labeling interactions as ‘positive’ or ‘negative,’” says Alhanai.
The above examples showing combination of AI and wearable components promises long-term positive growth for the healthcare industry. So it is just the beginning. The impact and possible applications of AI-powered wearable devices has immense potential and will continue to play a pivotal role in the healthcare sector.