E-HEALTH & DIGITAL TRANSFORMATION

HOLISTIC APPROACH: AI & AUTOMATION IN THE HEALTHCARE MANAGEMENT

The current situation and the challenges faced by the health community because of the pandemic, requires a rethinking and an optimization of the resources available, in order to respond to the citizens’ concerns.

In all sectors we have experienced an acceleration of digital transformation, and health is not a sector alien to this new digital reality.

In my humble opinion, ignorant and profane, the approach to health in a digital ecosystem must be holistic, a 360-degree digital transformation that may become a real lever for the optimization of available resources, to improve the experience of the citizen as patient when interacting with health professionals.

In 2018 I remember discussing the subject of robotics in an industrial and logistics environment with a friend, and in our conversation the interlocutor insisted on the term Cobot instead of Robot, which meant to explain the fact that the current ecosystem is no longer only human or only robot, but a kind of hybrid reality, with the interaction and collaboration between humans and machines, or robots, and this interaction meant a kind of collaboration and a symbiosis through digital transformation, this new symbiotic entity between the human being and the Robot, proves collaborative, hence the term Cobot is pertinent.

At the beginning of the pandemic, I expected to see intralogistic cobots, with onboard ultraviolet lights, to be used for cleaning and sterilizing hospitals, improving the user experience, mainly for patients with hypersensitivity, allergy or intolerance to detergents, solvents or similar, used for cleaning and sterilization of surgical rooms and sensitive areas inside hospitals; and definitively as we observed an increase in cases, manufacturers have been busy identifying these kind of new applications.

With a new paradigm, so-called digital, especially with the advent of autonomous vehicles, the range and spectrum of ecosystems using these cobots can increase ostensibly. An example would be the integration of force or pressure sensors, for example to weigh dialysis beds, and to define the time and treatments to be undergone in dialysis, according to the patient’s input weight, and finally to inform and document the data collected on the patient’s record file in the management and surveillance system, during the entire hospitalization of the patient.

Applications for Chatbot could still be found through the natural language processing #NLP, for example for general medicine, with a familiar Avatar, ensuring comfortable interaction with patients, which might be programmed to decide whether other humans are needed to validate the diagnosis and recommended initial treatments.

Using data collected from previously informed medical reports, Artificial intelligence could identify models and patterns, and through Machine Learning and decision trees, establish diagnosis and prescribe treatments; all based on preexistent recommendations and protocols established by leading medical associations and medical societies, so that their recommendations may feed a kind of database, where artificial intelligence is able to identify the best treatment suitable through its reconciliation with the symptoms identified in the patient.

The Internet #IoT is a layer that is also to be considered in the digital transformation, the implementation of sensors embedded in the physical environment, would also become a source of information and intelligence, through retrospective and prospective analysis; for example, portable devices to track cardiovascular events such as arrhythmia or hypertension or fever, and record them on the patient records either physical/paper or logic/SQL.

Identifying this kind of information may also become a first-line screening in the emergency room, since a computer will certainly process more field-related data than any human in a shorter time frame, and will report to doctors based on emergency levels, so that may improve the decision-making process, when prioritizing patients at the admittance at the emergency room.

Or integrate mobile solutions in hospital beds and chairs with sensors, depending on the medical specialty or the type of surgery or the patient’s clinics, for example, for an optimization of resources in post-surgical follow-up, where radiology staff for example can program the system, with an imperative command, so that the bed goes into radiology without human intervention, at the request of the staff, hence reduce the interaction with the patient. With the pandemic, the main concern for hospital staff and patients was the risk of contamination during the interaction, especially for patients with suppressed, depressed or compromised immunology.

On the intelligence and data management layer, the application of data management from a digital transformation perspective, proves that all the data collected during past and present pandemics, the availability and evolution of computation capacity through native or hybrid resources #Cloud, allows structuring all decision trees and algorithms, to feed a dedicated artificial intelligence #AI and begin an exercise of simulations, iterations and projections on the assessment of Future pandemics, and the strategy to follow in terms of action or prevention; not only in terms of planning events related to medical or health, but also to social or economic events, among other related indicators, and to identify the correlation. In short, re-write, record, inform, track, evaluate and understand the facts, patterns and models related to the main phenomenon or crisis; with the same goal: Predict, Prevent and Protect.

With the digitization of medical reports, Anonymization, its structuring with technologies like Folksonomy and Taxonomy in databases, and exploiting them by executing statistical tools, may enable identifying models and patterns in order to feed and train the AI with the results, to optimize its operation.

The Knowledge management and Transfer in the health sector is a must, and the professionals are aware of this, and it is for this reason that we may spot in the medical environment a common culture of learning, training and knowledge transfer through long life learning, formally or tacitly;  and the congress sector up to date has proven to be useful and essential for this task, nevertheless, the need for physical events and face-to-face conferences today can be replaced with a mixed reality, with virtual and augmented realities, not 100% but by reducing the weighing of the physical versus the virtual events, aimed at training and upgrading the professional knowledge and skills of the staff through new technological tools such as Share point or video conferencing type Zoom Teams or other. Or even through digital twins for simulations for training surgeons or with avatars and chatbots to train psychiatrists, for example.

Finally, what about the cybersecurity layer and related contingencies? how to secure data privacy and protect tangible and intangible assets in these strategic and sensitive infrastructures. With the Internet of things #IoT, we are increasing access to the infrastructure and endangering the entire ecosystem. Complex systems are vulnerable to their weakest link, and in digital transformation and industry 4.0, sensors and IoT can become the main threat to business continuity in the health sector and data privacy, since it is a potential source of digital Trojan horse.

Without digital sovereignty, there is no real sovereignty. Legislators have much to say about protecting intellectual and intangible assets and patient data privacy.

The principle “lex parsimoniae” advocates that “Pluralitas non is ponenda sine necessitate” or that plurality should not be postulated unnecessarily, according to my free interpretation, in digital ecosystems it is easier to inform and implement ML or DL, with decision trees and the optimization of Lagrange;  while humans, their decisions’ will potentially be contaminated by their ego and psychological state, and certainly influenced by their owns biases and cognitive distortions.

Maybe I am outdone by science fiction, didn’t Hume say that our imagination is limited by our knowledge? So long live to Star Trek and Star Gate.

To summarize, in my opinion to advance in the digital transformation in the health sector, it is necessary to ensure the digitization of patients’ medical records and reports in the first step, then appeal to descriptive statistics to initiate retrospective and proscriptive analysis, and to be able to exploit this data and create value for the entire chain of value. The second step in my opinion is Building Decision Trees and Algorithms that can empower the implementation of machine learning #ML, and this strengthened and fed with the information provided by the tools or sensors embedded in the hospital equipment. Next would be to look at the task of shaping the Artificial Intelligence, using the recommendations from the medical associations to feed it, to enable the AI to issue a potential diagnosis, prognosis and treatment based on the initial data collected, blood tests or imaging through sensors such as resonance or radiography. And finally with artificial vision and mixed realities, train Chatbots for example or Avatars, conceived to interact with patients, by its constant retro feeding with the results from the ML DL, to optimize the AI operation.

#healthcare #Artificiel #Intelligence #Tendance #Forecast #Taxonomie #Standaridzation

#BigData #Data #Business #Intelligence #Coronavirus

#Health #HealthCare #Management #Holistic: #Data #ML #DeepLearning #NLP #Chatbot #AI #Cloud #Computing #AV #Cobots #Folksonomy #Predict #Prevent #Protect #CSV #SQL

Ryan KHOUJA