Digital Healthcare Trends in 2022
Digital Healthcare Trends in 2022
Digital Healthcare Trends in 2022
The two-year-long COVID-19 pandemic has triggered a rapid development in digital healthcare. Demand for telemedicine has surged by 38 times, eHealth startups are receiving record funding, and medical institutions are investing heavily in their IT infrastructure. Sibedge identified the trends in the development of digital medicine that will be relevant in 2022 and invited Aleksey Moiseev, Ph. D. in Physical and Mathematical Sciences and the Head of the Department of Medical Physics at Medscan, as an expert.
Digital therapy (DTx) refers to mobile applications and online services that track the users’ state, help them establish a healthy lifestyle, get rid of bad habits, normalize nutrition and sleep. Most often, they offer exercises and practices that address physical and mental health issues. The main difference between digital therapy and other health devices and applications is the conduct of full-fledged clinical trials, and sometimes the necessity to obtain regulatory approval. Still, digital therapy does not replace traditional therapy but complements it. The advantages of DTx include high efficiency, no side effects, continuous monitoring of the patient's essential functions and no need to visit medical facilities.
The term "digital therapy" was first used in 2013 by the developer Sean Duffy to describe software that helps diabetic patients avoid disease through exercise and weight loss. A few years later, the FDA approved WellDoc, the first mobile diabetes app in the United States. It monitors the weight, physical activity, patient’s diet and transmits information to the attending physician who can adjust the treatment plan.
Digital Cancer Therapy
Applications for the physical and psychological support of cancer patients are in high demand. The study proved that patients who used meditation apps during an eight-week chemotherapy program experienced significant improvement in their emotional well-being. Another study demonstrated the effectiveness of digital therapy in cancer pain relief. The application continuously monitors the physical condition of the patient with the help of sensors, and if he is in pain, the doctor is informed about it. With the help of a remotely controlled micropump, the doctor injects an anesthetic drug into the patient's blood.
It is difficult to find useful and safe applications among several hundreds of thousands of those. This is one of the main problems of digital therapy. You can't trust user reviews on marketplaces or Google search results. People need guidance to be able to tell the difference between high-quality, regulatory-approved apps and low-end fakes. To help patients in their choice, German doctors have started prescribing medical prescription supplements (DiGa). Following Germany, France and 20 other countries are planning to join the new model.
“Nowadays, telemedicine offers great opportunities for both doctors and patients. That is why it has become one of the most promising areas in the development of the healthcare sector. Sibedge has participated in the development of a platform that allows remote patient visits by doctors and includes tools for monitoring their recovery. It improves access to quality health services by removing traditional barriers to health care, such as distance, lack of mobility, and time constraints. In situations where rural clinics do not have enough personnel or expensive equipment, telemedicine can save more than one life,” said Seva Morotsky, CTO of Sibedge.
“Previously, people were reluctant to resort to telemedicine. Now many have realized that it is fast, convenient, and can ensure high quality. It was with the arrival of the pandemic that the full potential of this technology was revealed. It is useful not only for patients, but also for medical professionals. All major scientific conferences are now online. Such events used to be exclusively face-to-face, but there was a forced shift. I hope this trend will continue,” said Alexey Moiseev Ph.D. in Physics and Mathematics and the Head of the Medical Physics Department.
Integration with Medical Equipment
Integration of telemedicine platforms with hardware is useful in several cases. For example, at the request of an American diagnostic company, we have produced a library that provides medical professionals with remote access to digital microscopes. The development of an API for remote access to devices that track patients' vital signs is always popular:
automatic blood pressure monitors;
blood sugar sensors.
Real-Time Video Communication
High resolution images are critical to ensure that the doctor does not miss any details during the remote examination of the patient. This is why our experts recommend using the following video compression standards for video communication:
AV1 — a promising new-generation open standard.
H.256/HEVC — better compression but increased load on the hardware platform.
H.264/AVC — acceptable compression and less load on the hardware platform.
Electronic Patient Records
To avoid confusion with a huge amount of information, it is important to document each meeting with the patient and attach audio, video, and text messages to the electronic medical record. This also allows you to bring the information to a standardized form. The most advanced platforms are closely integrated with the internal systems of medical institutions, combining records of personal visits and records of remote consultations into one map.
“During the quarantine period, working with medical documentation from home has become very popular. Now it is a big trend associated with the rationalization of the work of medical organizations. Digital medical records, for example, allow you to schedule the examination of patients in accordance with the schedule of doctors. With their help, it is much easier to track the revenue of a medical institution and keep statistics. This aspect will win the market in the future,” said Alexey Moiseev, Ph.D. in Physics and Mathematics and the Head of the Medical Physics Department.
Secure Data Exchange
All patient data transfer is carefully anonymized. Modern end-to-end encryption standards allow you to almost completely eliminate the possibility of information theft. Even so, there is a small chance of leaks. The diagnostic equipment has such a high resolution that using a special algorithm based on MRI and CT images of the head, it is quite possible to recreate a 3D portrait of the patient. This can be a serious problem if, for example, we are talking about celebrities.
“My colleagues from the UK have already highlighted this issue. Based on images of a person's head, you can recreate their portrait. Therefore, the transmission of such data in British medicine is prohibited. Work is already underway to hide the identity in images using artificial intelligence so that diagnostic information is not lost,” said Alexey Moiseev, Ph.D. in Physics and Mathematics and the Head of the Medical Physics Department.
Fast Image Transfer
Fast transmission of high-resolution images over the Internet is one of the cornerstones of telemedicine. When you need to quickly transfer tens of gigabytes of information from one clinic to another, the question arises of how to compress images quickly while keeping their high quality. The industry standard for data transmission, DICOM, is widely used in medicine. And the most advanced compression algorithms today are considered to be:
FLIF — the most efficient lossless compression;
WebP — algorithm from Google engineers;
Deflate — developed by Phil Katz, Creator of ZIP and PKZIP.
According to research company MarketsandMarkets, the market for medical software solutions based on artificial intelligence will be estimated at $45.5 billion by 2026. Despite the fact that the introduction of AI in healthcare is hindered by numerous regulatory barriers and concerns, this technology is gradually strengthening its position.
“It is quite difficult to implement artificial intelligence. And implementing AI quality control is a completely different level of understanding. I think we'll come to that, too. Due to all this hype many believe that AI is always right when it comes to diagnostics. Of course, it is not the case. One shall enter technical control of the software operation. It is an open niche, so quality control will be the next important step,” said Alexey Moiseev, Ph.D. in Physics and Mathematics and the Head of the Medical Physics Department.
German pharmaceutical company Boehringer Ingelheim together with Berg created artificial intelligence that accelerates the development of new drugs. And the American biotech company Moderna uses a neural network that works with matrix DNA and helps create a vaccine against COVID-19. In the future, this approach will allow us to quickly fight back against new viruses that threaten humanity.
Artificial intelligence has found application in radiation therapy. During the treatment procedure, you need to quickly get images, analyze them and make a decision: to continue or stop the procedure in order to correct something. Time for decision making is very short. AI allows you to speed up image processing and reconstruction, decision-making, and session adjustments right during the session.
“Calculating the propagation of ionizing radiation in a particular environment can pose a serious problem. My colleagues are using AI to speed up this process. First, the radiation dose is calculated at reference points, and then the AI builds a complete picture. The speed gain is huge. This can lead to a qualitative change in medical processes. Previously, radiotherapy planning was performed by a person, but now it can be replaced by a computer. Of course, adjusted for the already mentioned quality control,” said Alexey Moiseev, Ph.D. in Physics and Mathematics and the Head of the Medical Physics Department.
Artificial intelligence training is impossible without input data standardization. Names, types of research, and pre-prepare images should be unified. Today, this process is a serious issue for AI developers. Complex anthologies are introduced, and all data is broken down into specific models to make them suitable for AI training. Lack of standards results in a problem with big data and, consequently, with AI training.
“We are seeing a global trend in the desire to standardize everything and everything. Without standardization, AI development is impossible. When the source data is bad, the result of the AI will be bad as well. Qualitative data equals standardized data. I hope this trend will continue,” said Alexey Moiseev, Ph.D. in Physics and Mathematics and the Head of the Medical Physics Department.
Image Analysis and Diagnostics
IBM continues to improve its supercomputer which is already independently make accurate diagnoses for patients and select effective treatment methods. Artificial intelligence has learned to analyze anamnesis and medical images (radiography, MRI, CT, etc.), as well as to find pathologies that the doctor might have overlooked with high accuracy. Such a system is trained on the basis of huge amounts of data, which is a long and expensive process.
“Artificial intelligence processes the available data and passes it to the doctor for verification. This is called decision support. AI focuses the doctor's attention on certain important details, which makes the process much faster. The final decision is still made by the doctor, but the process is partially automated,” said Alexey Moiseev, Ph.D. in Physics and Mathematics and the Head of the Medical Physics Department.
The use of existing drugs for alternative indications significantly reduces the cost of creating completely new drugs. Artificial intelligence analyzes the effects of the drug on the body, takes into account side effects, and then simulates situations in which, for example, you can use a drug for malaria to fight other diseases. Since these drugs have already been approved by regulatory authorities, the approval process for the second indication is much faster.
Healthcare often depends on complex and extremely expensive research. High technologies can not only facilitate the work of laboratory staff but also significantly speed up these processes.
Analysis and mathematical modeling of disease outbreaks play a very important role in planning the response of health authorities to infectious disease outbreaks, epidemics and pandemics. Major IT companies are currently working on creating neural networks that can predict how quickly and in what manner new diseases will spread based on the analysis of relevant data. This will allow humanity to better prepare for new threats.
Telemetry Clinical Trials
In the context of the COVID-19 pandemic, pharmaceutical companies were forced to suspend clinical trials of medicines and new vaccines. The subjects simply could not be present in the laboratories. Remote monitoring of control group participants was introduced as an alternative to traditional research. Software solutions such as eCOA (Electronic Clinical Outcome Assessment) and ePRO (Electronic Patient-reported Outcome) are actively used by market leaders today.
Operation with big data allows you to analyze huge amounts of information. These studies may result in lower treatment costs, early screening of a number of diseases, and improved quality of life in general. Today, data science is used in solving global issues such as finding cures for cancer, HIV, and other dangerous diseases.