Personalized treatment (PT) is an emerging area in healthcare that provides personalized health. Personalized, targeted, or customized treatment gains more attention by providing the right treatment to the right person at the right time. Traditional treatment follows a whole systems approach, whereas PT unyokes the people into groups and helps them in rendering proper treatment based on disease risk. In PT, case by case analysis identifies the current status of each patient and performs detailed investigation of their health along with symptoms, signs, and difficulties. Case by case analysis also aids in constructing the clinical knowledge base according to the patient's needs. Thus, PT is a preventive medicine system enabling optimal therapy and cost-effective treatment. This chapter aims to explore how PT is served in works of literature by fusing machine learning (ML) and artificial intelligence (AI) techniques, which creates cognitive machine learning (CML). This chapter also explores the issues, challenges of traditional medicine, applications, models, pros, and cons of PT.
TopIntroduction
Personalization is a multidisciplinary research topic which triggers the computational intelligence in tailoring specific services to each person. In clinical practice, PT helps in developing treatments that are specific to an individual or group of individuals. PT enhances better patient care and improves healthcare by providing targeted therapy for each person. PT is a case-based model for effective decision making. Tailoring precise treatment to the precise patient at the precise time is the fundamental constituent (Xie Q et al., 2019) of PT. In recent days, AI, ML and Deep Learning (DL) have sparked interest in augmenting medical decision making. ML and AI include brain-inspired algorithms to achieve individual-level clinical predictions. A recent report estimates that a sharp increase of nearly $87 billion in personalized medicine technologies has been expected by 2023 (Aguado et al., 2018). PT treats the patients individually by coupling their genetic information with medical records like Electronic Medical Records (EMR), Electronic Health Records (EHR), and Personal Health Records (PHR). EMR is a computerized version which includes patient's medical history, immunization status, medications, laboratory and test results, diagnoses, progress notes, patient demographics and vital signs. EHR is a comprehensive snapshot of detailed patient's medical history. PHR includes family medical history, prescription records, observations of daily living, chronic diseases, illness, and hospitalizations. Clinically oriented decision-making system is developed for breast cancer patients (Jiang et al., 2019), which provides customized assessments and recommendations by accumulating their health records. This decision support system makes use of Bayesian network architecture and Treatment Feature Interactions (TFI) algorithm to provide optimal and individualized treatment decisions.
A digital oncology platform is created for cancer patients (De Regge et al., 2019) portraying the personalized care path. The care path contains a customized pathway that provides righteous information about treatment options of cancer, pointers to related websites, questionnaires and treatment suggestions from the medication team. Using this care path, doctors can easily personalize and choose the necessary treatment for the patients. A cognitive computing technology-based medical recommender system (Dessi et al., 2019) IBM Watson and Framester performs feature extraction, processes the digitized patient's records and clusters a pile of medical reports. Given, a patient's medical report, the system can detect cohorts of patients with the homologous symptoms. The CML combines the techniques of AI, ML, DL and NLP to provide targeted therapy. CML proves to be an optimal environment for providing PT since it includes the advanced ML and AI techniques. CML helps the doctors for analyzing the massive volume of unstructured medical records efficiently. Doctors usually depend on their experience in diagnosing the illness and the review process is time-consuming, whereas the CML technology saves time by inferring the correct medication from the analysis. Figure 1 depicts the outline of PT for similar cohorts of patients.
Figure 1. Outline of personalized medication using cognitive machine learning
Firstly, the CML framework analyzes the patient's records by incorporating the techniques of AI, ML, DL and NLP. Secondly, the framework constructs a clinical database and provides sufficient information about the diseases. Thirdly, the framework involves the identification of cohorts of patients to deliver personalized medications. PT is used to determine the most optimal therapy by an accurate assessment of a patient's condition. This chapter explores the following objectives
- •
Different approaches in ML and AI for providing personalized patient care.
- •
How Cognitive Machine Learning (CML) techniques provide an optimal treatment strategy for each patient?
- •
How PT shapes the future of healthcare?