Image credits: Wikimedia Commons

AI Creating a Virtuous Circle for Health Systems

By Miriam Bello | Mon, 03/08/2021 - 09:46

The digital transformation in healthcare has just accelerated as a result of COVID-19 resulting in benefits for patients toward a sustainable health system in the long term. Artificial Intelligence (AI) has been one of the main protagonists in Healthcare 4.0 due to its positive impact on businesses and care provision.  

AI for healthcare has many applications, including:

  • Healthcare personalization
  • Diagnosis and treatment applications
  • Patient engagement and adherence to treatment
  • Administrative activities
  • Health system unification
  • Mental health crisis prevention
  • Access to care in underserved or developing regions

Health systems and individual companies are starting to invest in AI with the goal of growing process efficiency (34 percent), enhancing existing products and services (27 percent) and lowering costs (26 percent), according to a Deloitte survey.

AI applications can also create a virtuous cycle. For instance, healthcare personalization through machine learning allows an appropriate prediction of treatments, which can be backed up by a trained database of AI systems. This also allows a standardization of processes, which can lead to cost and time savings for both care providers and patients. To make a correct prediction, AI systems take into consideration genetics, diet, lifestyle, socioeconomic status and external risks factors. 

Aidicare is a company that has integrated this concept into a service called Aidicare Life. During an MBN interview, CEO Juan Cáceres explained that this product is designed to help patients care for their health outside the hospital through a wearable and an app that track the person’s vital signs, including sleep patterns, physical activity, daily steps, weight control, glucose control, heart rate, blood pressure, oxygenation and temperature. “To connect this information with a doctor and offer a complete healthcare service, Aidicare also provides a digital board where doctors can keep track of patient’s health,” said Cáceres. This customized solution using AI algorithms allows for a precise diagnosis and advice for a healthier lifestyle. The objective is to provide doctors with a clear view of the patient’s lifestyle, habits and health status, which also saves around 35 minutes during regular consultations.

Similar to personalized healthcare, diagnosis and treatment applications can be supported by an AI-powered database. However, their precision is still being perfected as, according to NCBI, “they lack the precision of more algorithmic systems based on machine learning.” Rule-based clinical decision support systems are difficult to maintain as medical knowledge changes. Moreover, they “are often not able to handle the explosion of data and knowledge based on genomic, proteomic or metabolic approaches to care,” states NCBI. The latter has represented a barrier to the full integration or investment in AI systems. Nevertheless, many tech companies (rather than health companies) have found applications in some medical fields, such as radiology, which derived on the current trend called radiomics. AI has also proven to be a supportive tool for physicians as a “quality control tool” or a first filter to detect a red light on a patient trough their habits.

Patient engagement and adherence to treatment is described often as the “last-mile” for healthcare efforts and what determines healthcare outcomes. Beyond patients’ participation in treatments, their involvement in self-care trends can have a more positive outcome. AI can offer support through personal devices, setting reminders and tracking to ensure and generate attachment to treatments. This can provide more information to care providers so they can adapt their recommendations for an effective treatment.

Administrative activities are focused on improving internal performance for care providers. Their main benefits can be seen in hospital functionality, ensuring companies make a better use of their time and have further control over their supplies. “At CHRISTUS MUGUERZA for example, doctors use a computerized physician order entry (CPOE) to prescribe a treatment for a patient,” said Jesus Díaz, CIO of CHIRSTUS MUGUERZA. Díaz explained that this system has standardized measurements of medicine and times to avoid mistakes in care provision, while making it easier for other divisions of the hospital to be aware of the patient’s situation. Communication between nurses and doctors is key in every process and by having digitalized information, they can easily be on the same page regarding the patient’s situation. According to Díaz, all these processes are patient-centric and are created to offer personalized services.

Other companies develop more specialized solutions, like Aidicare. Cáceres told MBN how the company developed a solution for ICU units at hospitals. “It uses a tablet to track vital signs and alert medical staff of any changes in the patient’s condition. Through the use of AI, Aidicare offers facial recognition of the patient before entering the ICU or before surgery, which is possible through a partnership with Amazon Web Services,” he said. The objective of this development is to avoid any misdiagnosis or use of a wrong clinical record.

Moreover, insurers can also benefit from this, as they have a duty to verify whether millions of claims are correct. According to NBCI, “incorrect claims that slip through the cracks constitute significant financial potential waiting to be unlocked through data-matching and claim audits.” During COVID-19, AI has been a key tool for insurers. Cáceres explained to MBP that from Aidicare’s joint work with Médica Sur, the company created an option for insurers to provide real-time follow-up on one of their clients who is a confirmed case of COVID-19. This saves time and costs for both the hospital and the insurer.

Unification of health systems might be one of the most valuable aspects of AI for public healthcare. Currently, the main goal for this industry is to move towards “Value-Based Healthcare” but to make that happen countries need to have an integrated health system, which, according to a statement made by Diego Guarin, Regional Market Access Lead for Latin America at Merck MSD, one of the first steps toward a migration to this system in both the public and private sectors is the development of tools or technology platforms for coordinated patient follow-up.

Altron Healthtech states that AI helps to make Value-Based Healthcare possible by developing growing understanding of data. An MBN analysis describes how technology and AI for health systems helped with resources calculation and ECR integration to help with the overall interconnection of the system and data exchange among public and private entities.

Through its algorithms, AI is also able to prevent or predict mental health crisis like suicides or overdoses. There are in fact apps already dedicating to this such as Yana, a Mexican company which developed a mental health chatbot that uses AI recognition methods and natural conversation to help users tackle depression and anxiety. Andrea Campos, Founder and Director of Yana, explained to MBN that while this is not a crisis app, “our application does identify speech that can indicate a crisis. The application includes an SOS button, which directs you to the government’s suicide hotline.” According to Healthcare Innovation, “AI can also help non-behavioral health providers understand how mental illness increases patients’ risk for other negative outcomes such as lack of exercise or bad eating.”

Expanding access to care in underserved or developing regions might be one of the most valuable contributions for countries like Mexico, where the lack of infrastructure and medical professionals has been a historic barrier for health provision. In this area, Nubix is an app addressing one of Mexico’s major problems: radiologists’ limitations. Nubix is a digital platform that maximizes the diagnostics capacity of remote facilities due to their geographical location or lack of professionals. Through the app, general doctors can update patient information and have a response within a maximum of two days, compared to the traditional two months for a single radiology interpretation.

Looking at the bigger picture, AI has made telemedicine possible through the integration of services like radiomics, ECR, electronic prescription and patient monitoring. Telemedicine has been identified as an effective way to reduce inequality to healthcare access, as long as there is commitment from different actors. During an MBN interview, Alessio Hagen, Director of Digital Cities for Latin America at Dell Technologies, addressed how leaders of this industry can effectively bring telemedicine to rural areas for Mexico, which are the ones with less healthcare access and less technology access. Hagen mentioned that Dell has taken action to address both problems. “We introduced rural telemedicine using pictures of the children’s ailments to ensure early cancer detection.” Dell was able to provide telemedicine centers where an oncologist could treat children by using a computer. The consultation was supported by a general doctor and the child’s parents. “As this was implemented in rural areas with no internet, we used satellite connection,” Hagen explained.

Mexico Adopting AI

“AI and machine learning do not exist without solid databases,” said Mario Muniz, Regional General Manager for North Latin America of IQVIA, to MBN. “In Mexico, the health system is fragmented and it is difficult to connect the databases in the public and private sectors. One day a patient may go to one place, the next day they might go to another and their medical record is nontraceable, nor uniform.” Muniz noted regulation is the second factor acting as a barrier for digitalization. “Proper regulation could boost our solutions to provide better data that can generate adequate solutions and derive in costs savings. Mexico is a fertile soil for technology innovation. However, cultural changes will allow it or prevent it.”

Hagen highlights another big obstacle. “I think resistance to change has been the main obstacle. We need to transmit that the solutions are created to simplify their process, while also listening to feedback to create products that really support doctors’ practices.” While Mexico is still far from completing its digital transformation and the unification of its system, Hagen told MBN that Dell has been working with the public sector on the interconnection of institutions, which is seen as a positive step to think about a future connected system.

To close the virtuous circle that AI can provide, basic management is required. “Without it, the use of basic technology will never be encouraged and digital solutions will forever be a neglected tool,” said Mario Sicilia, Director General of Laboratrio Médico Polanco, during Mexico Health Summit 2021.

Photo by:   Wikimedia Commons
Miriam Bello Miriam Bello Journalist and Industry Analyst