AI and machine learning in healthcare: now is the right time to invest

AI and machine learning in healthcare: now is the right time to invest


While discussing the economic sustainability of the sector, Saúde already finds practical examples against waste in technology




AI and machine learning in healthcare: now is the right time to invest

More accurate diagnoses, increasingly rapid and accessible therapeutic treatments, creation of assertive health protocols and development of personalized and more effective drugs. These are just some of the benefits that artificial intelligence (AI) can bring to medicine, as it is able to identify meaningful relationships in raw data – such as medical reports, medical records and clinical trials – and to cross-reference this knowledge with the data of the population. .

In healthcare management, the insights that can be revealed by artificial intelligence and machine learning are of equal importance. However, there are still few healthcare players that are tapping all this potential into new businesses.

Those who have already joined this digital transformation are now able to manage resources more securely, without compromising the quality of the service offered. This is precisely the case of some national healthcare companies which are eliminating the characteristic waste of this sector, such as the duplication of exam requests, with the help of information.

“Startups offering affordable consultations and digital health plans are some examples of new businesses that have been facilitated and empowered by the use of data,” says Paulo Fernandes Silvestre, master in intelligence technologies and digital design, who adds: “ Often, the use of AI and machine learning is not in healthcare, but in healthcare management, such as to reduce fraud.”

“Since AI is the technology used to identify patterns, when applied in audits and medical bill tracking, this feature has brought great results,” reveals the expert.

In addition, AI-enabled systems can anticipate overloaded care beds or drug shortages, which enables hospital management and even public health management to make proactive decisions. On the healthcare front, there are already studies that calculate the reduction in treatment costs from the personalized measurement of the use of chemotherapy drugs. In this way the patient receives a less toxic dose and with the same efficacy expected from his treatment.

The possibility is not lacking

In the area of ​​signals and images, there are even more applications already available, such as Renato Sabbatini, director of Digital Health Education at the Associação Brasileira CIO Saúde (ABCIS) and one of the pioneers of Health Informatics in Latin America , recalls: “This is the case of the automatic diagnosis of the ECG, which promotes the detection of atrial fibrillation and the processing of signals generated by patients in wearable devices, such as smartwatches. As for image processing itself, although systems based on machine learning are registering high efficiency, the truth is that they are still far from being used in routine radiology, perhaps because their usefulness is still rather limited”.

Another application area that has developed rapidly and is widely adopted is NLP – Natural Language Processing -, with text-to-speech and voice-to-text conversion and intelligent chatbots, although, in many cases, the use of this type of technology is still extremely inaccurate and, at times, dangerous when aimed at patient care.

“There is a lot of ‘hype’ in this type of technology, as was the case in the previous waves, which ‘winters’ followed. Despite the great clamor, there are still few applications of AI and ML for routine use in clinical practice”, clarifies Sabbatini. And this is also where the opportunity lives.

Although AI and ML technology are already present in Health, it is a fact that there is still a lack of professionals who understand both care and management and the business itself. By working closely with developers, they should be able to optimize solutions that actually help the industry thrive.

“Apple, SAP and IBM are examples of technology companies that are already investing in health care, even if their primary business is elsewhere. It remains for health professionals to take the opposite path – i.e. invest in more technology – so that the two sides meet halfway in the near future,” recommends Paulo Silvestre.

Is investing still expensive?

Advanced AI and ML systems depend on high computational capacity and investment, that’s clear. And knowing this can make many institutions fear crossing this technological barrier to bet on innovation – which no longer makes sense given that the technology business model has changed, as Sabbatini clarifies: “The competition from AWS, Microsoft, Google and other bigtechs to offer cloud computing services for the areas of AI and ML, such as Tensor Flow, have made the costs more affordable. In addition, software and database tools for hyperparallel processing, as well as virtually unlimited Web storage, are leveraging many technology initiatives that would have been cost-prohibitive in the past.

Medicine, after all, is art.

Even with so much innovation and a desire to change the status quo in the industry, there is still a risk that new businesses will run into cultural issues, such as lack of training of healthcare professionals or even the patient’s preference for the human touch of the doctor instead of the analyzes performed by the “machine”: “Hippocrates already said that medicine is not only science, but also art, the Ars Cvrandi, in Latin. The training of professionals in the use of these technologies is essential, however we do not yet have enough courses, disciplines and teachers to introduce the subject into undergraduate and graduate curricula”, warns Sabbatini.

However, health technology regulatory bodies, such as the National Health Surveillance Agency (Anvisa), are already making progress in these aspects, as evidenced by the recent RDC 657/2022 regulation on software as a medical device (SaMD), which will be a great challenge for all solution developers for years to come.

While AI and ML applications have leveraged solutions in healthcare, largely thanks to health technology initiatives and/or tools made available by big technologies, ethical and cultural barriers are still imposed and require more discussions by all the actors involved. After all, now is the right time to invest – even more – in this type of solution.

Renata Armas is the editor ofunbox

.+

The best content in your email for free. Choose your favorite Terra newsletter. Click here!

Source: Terra

You may also like