The symbiotic growth of machine learning (ML) and artificial intelligence (AI) is expanding the value of the Internet of Medical Things (IoMT). Physicians are able to draw actionable conclusions more quickly and reliably when processing massive, continuous streams of data from connected medical devices.
What is the Internet of Medical Things (IoMT)?
With IoT use cases proliferating across industries, the medical field is no exception. Metrics and patient feedback have become essential for healthcare providers as U.S. healthcare transitions toward evidence-based outcomes and incentives begin to align. Payers are increasingly looking to optimize costs with treatments that are more effective than alternatives.
Data-based care will continue to grow and will have a beneficial impact on the cost and quality of care.
The key value that sensors bring to healthcare is reducing the time between measurement, detection and treatment. Insulin pumps measure and deliver doses at the appropriate times using a blood sugar monitor that has sensors under the skin that communicate blood sugar levels to an external receiver. Additionally, data analysis capabilities are now available to add context and meaning to measurements faster than ever before.
Taxonomy of Medical IoT Use Cases
Diagnosis
How Can Medical IoT Devices Improve Diagnosis? Devices may increasingly track body markers that can indicate medical conditions such as diabetes and atrial fibrillation. Key medical parameters such as blood chemistry, blood pressure, brain activity and pain levels can be collected continuously.
This could help detect early signs of disease onset or activity, leading to improved responses. Once disease susceptibility or risk factors have been identified, causal indicators can be closely tracked with the right targeted sensors. Even the latest version of the Apple Watch 4 has been declared a Class 2 medical device thanks to features like heart rate monitoring and fall detection.
It must be noted that most consumer-oriented devices have not yet passed the FDA regulatory process and cannot be considered medical devices.
Rehabilitation
Patient recovery time after surgery is an important component of surgery costs, and minimizing surgery time is an important factor in reducing costs. For example, for a total knee replacement the length of hospital stay in the US is about two days, compared to four to five days in the National Health Service (NHS) outside the UK, requiring reduced hours in SNFs (skilled nursing facilities) and physiotherapy. This can be achieved through the use of wearable sensors that aid in exercise, compliance and remote monitoring.
Sensors can track various key indicators and alert caregivers to respond in a timely manner. Sensors combined with telemedicine make it easier to speed up recovery. Knowing what a patient is doing between visits can help shorten recovery time after surgery. In fact, a three-year collaboration between the Geisinger System and Force Therapeutics has led to significant improvements in treatment outcomes. This includes a 30% reduction in hospital stays, a 56% reduction in skilled nursing facility use, and an 18% reduction in readmissions, reports Greg Slabodkin of Health Data Management.
Long-term care
Sensors that track bodily parameters are becoming increasingly sophisticated, with advances in the analysis of blood pressure, glucose levels, sweat and even tears. The benefits of this compared to standardized testing lie primarily in the frequency of data capture. In the case of chronic degenerative diseases such as rheumatoid arthritis, movement sensors can help improve gait and form it. Another class of IoMT device applications is monitoring and responding to patient adherence to treatment. Especially in chronic care, adverse outcomes and prolonged recovery periods can be avoided through measurement and monitoring ideally suited for IoT devices.
Precautions
Devices that actively engage patients in guided exercise can help prevent the need for medical attention and associated costs. For example, the range of motion of joints in the orthopedic space, or postural adjustments to prevent cervical spondylosis, are examples of how devices can help with prevention. An example is Upright.
For example, wearable devices can prevent falls in the elderly by checking their activities and noticing any anomalies that may lead to imbalance and falls. Apple's Watch uses a built-in IMU (inertial measurement unit) to identify a drop or possibility. It could even be used to measure tremors associated with neurological disorders such as Parkinson's disease.
Conclusion
Medical IoT is increasing in value due to the symbiotic growth of machine learning (ML) and artificial intelligence (AI). When processing massive continuous streams of information from sensor-assisted medical devices, data analytics and ML can provide actionable conclusions faster to aid in the healing process.
Through the flow of information, preventive care can reduce hospital admissions and significantly lower emergency department costs. This will increase efficiency and improve patient satisfaction and outcomes. However, some risks to data security in motion and at rest must be carefully assessed. Additionally, the risk of false-positive readings creates unnecessary stress for patients and care systems. Accuracy, repeatability and reliability are the three essential elements of IoMT, which must always be prioritized.
