Health care and service.

 

Healthcare services have seen a strong paradigm shift in recent years globally. At the clinical level, personalized care to individuals is usually provided based on medical history, examination, vital signs, and evidence. However, in the recent times, the focus on these traditional tenets is being taken over by the aspects of learning, metrics, and quality improvement [1]. 

The last decade has seen a global rise in adoption of Electronic Health Records⁹ (EHRs) [2–7], catalyzing the increase in the complexity and volume of the health data generated in the process. Apart from the EHR-sourced ordinary patient data, due to the change in treatment paradigms and focus on lifestyle and comprehensive healthcare, new varieties of health-data about medical conditions, lifestyle, underlying genetics, medications, and treatment approaches also showed a paramount rise.

Despite the complex nature of new-generation health data, human cognition to analyze and make sense of these humongous data is finite [8]. The traditional medical models of analysis deserve a reengineering for more efficiency, leading to the computer-assisted methods to organize, interpret, and recognize patterns from these data [9, 10]. Efficient collection and accurate analysis of data are critical to improvements in the effectiveness and efficiency of healthcare delivery [1]. In this respect, emerging as a promising field, eHealth addresses multifarious aspects of the healthcare system, like tracking changes in health behavior and prevention and management of chronic diseases [11].

In the recent years, the intrinsic power of data in healthcare started unveiling like never before, leading to the endeavor in making sense of health data in the best possible way, using advanced data analytics and computational intelligence. Especially in the field of healthcare, the aspect of intelligent data analytics is one of the most trending topics worldwide [12]. One of the prime areas where such analyses have been applied is the field of chronic diseases. By 2020, chronic diseases are expected to contribute to 73% of all deaths worldwide and 60% of the global burden of disease. 

Also, 79% of the deaths attributed to these diseases occur in the developing countries. Four of the most prominent chronic diseases—cardiovascular diseases (CVD), cancer, chronic obstructive pulmonary disease, and type 2 diabetes—are linked by common and preventable biological risk factors, notably high blood pressure, high blood cholesterol, and overweight, and by related major behavioral risk factors. To prevent these major chronic diseases, the actions need to be centered around controlling the key risk factors in a comprehensive and integrated manner [13]. In addition to the chronic diseases, the aspect of intelligent risk prediction and preventive actions count significant even in the domain of transplantations [10, 14].

Moreover due to the concept of context awareness backed by sensor fusion in the environment of smart eHealth systems and IoT, the health data generated and acquired is more comprehensive and detailed. The smart prediction and prevention systems in healthcare usually share some common steps like the collection of health data from sensors or other sources and assimilating the EHRs, followed by analyzing and computing the risks [15]. In the endeavor of taking possible actions to prevent chronic diseases, detecting the diseases at an early stage stands to be the prime challenge. Most of these diseases do not exhibit clearly identifiable signs at the early stage. And here lies the key area of Artificial Intelligence (AI), harvesting the possibility of early detection of these diseases in terms of risk. 

From the data science perspective, the aspect of health data acts as the key valuable resource. Most importantly, the domain of health data has expanded dramatically over these years. Even the superficially and noninvasively obtained behavioral, physiological, and metabolic health data hold enormous significance. In the domain of early detection and prediction of diseases, the health data possesses a huge potential. Disease prediction led by AI is a multilevel process. It involves the analysis of the intricate details inside the health data, looking for early indications or traces of diseases.

Thanks to the recent boost in paradigms like IoT, eHealth, and medical informatics pertaining to AI, healthcare is one of the most important areas where data analytics finds its applications and analyzing health data has reached new heights [15a]. Minimizing the response time in diagnosis and treatment is a crucial component in efficient healthcare services, which makes the power of data analytics relevant for faster analysis and use of intelligent methods for better diagnosis [16]. Detailed analysis of health data expedites the automated diagnosis on one hand, also leading to personalized treatment.

 On the other hand, it provides the comprehensive and holistic information of a large group of people under treatment. This is fairly advantageous to automate the process of monitoring along with prediction of health risks obtained from the analysis of the health data of the patients. In this direction, one of the key areas where prediction of risk is highly crucial is transplantation. 

Transplantation is itself a complex procedure that makes the consideration of pretransplant predictions of risk an important factor, opening up a broad scope for computational intelligence. Possible insights obtained from the pretransplant health data of the patients by harnessing the power of AI would positively influence the healthcare delivery approach. In this work a case study is presented highlighting the aspects of computational intelligence toward risk prediction in liver transplantation.

Liver transplantation is the last therapeutic option in patients with end-stage liver disease. Being a complex healthcare process, it is related to humongous costs and requires the expertise of a specialized interdisciplinary team along with a close monitoring of patients during the entire timeline. This process generates a large volume of complex, multidimensional data. The adequate clinical management of transplant patients impacts their vital prognosis, and decisions on many occasions are made from the interaction of multiple variables [17]. However, there exists an enormous demand in the domain of prediction in liver transplantation process, ranging from survival till the suitability of transplant [18].

The healthcare sector has emerged as one of the prime areas to adopt new technologies, given that the primary objective is to provide better and more efficient treatment to the patients. Healthcare delivery is a complex aspect at both individual and population levels. From the perspective of data-driven insights for any medical personnel, a large pool of historical health data of the patient is a huge plus, before starting a thorough treatment [19]. At the clinical level the aspect of providing healthcare services is guided mostly by medical history, examination, vital signs, and evidence. 

Recent times have seen a paradigm-shift of the core traditional approaches toward the supplementary focus on learning, metrics, and quality improvement of the healthcare provided. The collection and analysis of good-quality data are critical to improvements in the effectiveness and efficiency of delivering healthcare services [20]. The field of artificial intelligence applies to a wide range of disciplines in medicine; however, in transplantation, it is still a scarcely explored area.

The AI has started playing an important role in predicting the main determinants of morbidity and mortality in patients, which stands quite significant in the domain of transplantations. This deals with analyzing the probability of developing an inherent risk of disease or complication during the entire timeline. 

The main objective of this work is to spotlight the applied aspects of data analytics in healthcare and importance of AI in transplantation, illustrated through a case study of liver transplantation at the National Center for Liver Transplantation and Liver Diseases, Uruguay. Also, based on the advantages of AI in transplantation, an AI-based predictive clinical decision support system for transplantations has been proposed for early detection and prediction of risks and proffering better diagnosis and treatment to the patients.