Generative Artificial Intelligence-based Treatment Planning in Clinical Decision-Making, in Precision Medicine, and in Personalized Healthcare.

• Author: Grupac, Marian

1. Introduction

   ChatGPT can predict diagnoses, enable virtual consultations, recommend specific treatments, efficiently inspect massive datasets (Dabija et al, 2018; Nica, 2018; Pelau et al, 2021; Vatamanescu et al, 2022), monitor health data and patient care, and lower healthcare costs. The purpose of our systematic review is to examine the recently published literature on generative artificial intelligence-based treatment planning and integrate the insights it configures on clinical decision-making, on precision medicine, and on personalized healthcare. By analyzing the most recent (2023) and significant (Web of Science, Scopus, and ProQuest) sources, our paper has attempted to prove that ChatGPT can provide well-structured, relevant, and accurate medical information as regards patient education and support through personalized advice while requiring optimized support system reliability and applicability. The actuality and novelty of this study are articulated by addressing how ChatGPT can assist physicians and patients by inspecting complex datasets and producing personalized treatment recommendations, that is an emerging topic involving much interest. Our research problem is whether generative artificial intelligence algorithms (Andronie et al., 2021b; Musova et al, 2021; Popescu et al, 2017a) can be deployed as regards laboratory diagnostics, public health outbreak management, clinical decision support tools and trial data, patient care efficiency, and drug discovery.

   In this review, prior findings have been cumulated indicating that by accessing patient medical records, generative artificial intelligence algorithms (Andronie et al, 2021; Naepi and Naepi, 2022; Popescu et al, 2017b) can improve clinical imaging and reasoning, identify medical symptoms, and consolidate postoperative management and rehabilitation. The identified gaps advance how ChatGPT produce factually precise and contextually significant structured discussion feedback (Andronie et al, 2021a; Duncan, 2022; Nica et al, 2022) to elaborate and emerging clinical questions, swiftly retrieving and integrating complex medical information into streamlined answers. Our main objective is to indicate that generative artificial intelligence algorithms (Andronie et al, 2023; Nica, 2017; Peters et al, 2023; Popescu, 2018) can streamline clinical workflows, provide quite error-free medical information, and enable optimal patient care and clinical decision-making.

2. Theoretical Overview of the Main Concepts

   ChatGPT provides quite appropriate and readable medical knowledge with respect to diagnostic methods, visual prognosis, and surgical and nonsurgical treatments in terms of health-related counseling and factual information for both patients and physicians. Machine and deep learning-based generative artificial intelligence technologies integrate patient medical history, personalized health information, symptoms, and laboratory outcomes. Generative artificial intelligence algorithms can improve the communication and knowledge exchange of accessible and understandable health information and literacy, furthering users in making accurate informed decisions. ChatGPT can harness image analysis tools adequately, interpret clinical cases precisely, and evaluate treatment progress, resulting in enhanced personalized care. The manuscript is organized as following: theoretical overview (section 2), methodology (section 3), ChatGPT can provide well-structured, relevant, and accurate medical information (section 4), ChatGPT can predict diagnoses, enable virtual consultations, and recommend specific treatments (section 5), generative artificial intelligence algorithms can streamline clinical workflows (section 6), discussion (section 7), synopsis of the main research outcomes (section 8), conclusions (section 9), limitations, implications, and further directions of research (section 10).

3. Methodology

   Throughout April 2023, we performed a quantitative literature review of the Web of Science, Scopus, and ProQuest databases, with search terms including "generative artificial intelligence-based treatment planning" + "clinical decision-making," "precision medicine," and "personalized healthcare." As we inspected research published in 2023, only 188 articles satisfied the eligibility criteria. By removing controversial findings, outcomes unsubstantiated by replication, too imprecise material, or having similar titles, we decided upon 40, generally empirical, sources (Tables 1 and 2). Data visualization tools: Dimensions (bibliometric mapping) and VOSviewer (layout algorithms). Reporting quality assessment tool: PRISMA. Methodological quality assessment tools include: AXIS, Dedoose, MMAT, and SRDR (Figures 1-6).

4. ChatGPT Can Provide Well-Structured, Relevant, and Accurate Medical Information

   Machine and deep learning-based generative artificial intelligence technologies can configure an accurate diagnosis (Chavez et al., 2023; Grunebaum et al., 2023; Kim et al., 2023; Theodosiou and Read, 2023), optimize health service quality towards precision medicine and personalized healthcare and analyze medical images. Generative artificial intelligence algorithms can be deployed as regards laboratory diagnostics, public health outbreak management, clinical decision support tools and trial data, patient care efficiency, and drug discovery.

   ChatGPT can ensure content accuracy, clarity, readability, and coherence (Abd-alrazaq et al., 2023; Lee, 2023; Mohammed et al., 2023; Temsah et al., 2023; Vaishya et al., 2023) in teaching and learning processes associated with medical education through streamlining data extraction and summarization, writing style and formatting support, and subsequent vast amounts of first-rate work production. Machine and deep learning-based generative artificial intelligence technologies integrate patient medical history, personalized health information, symptoms, and laboratory outcomes.

   ChatGPT can assist physicians and patients by inspecting complex data-sets and producing personalized treatment recommendations (Ge and Lai, 2023; Johnson et al., 2023; Liu et al., 2023; Seth et al., 2023; Wagner and Ertl-Wagner, 2023): massive genomic and clinical data are pivotal in accurate treatment outcome predictions, exemplary therapeutic approach determination, and clinical trial matching assistance. ChatGPT can provide well-structured, relevant, and accurate medical information as regards patient education and support through personalized advice while requiring optimized support system reliability and applicability. (Table 3)

5. ChatGPT Can Predict Diagnoses, Enable Virtual Consultations, and Recommend Specific Treatments

   ChatGPT can generate a coherent and brief clinical history by data source summarization, incorporating prior imaging reports and consequently producing an informed and plausible report while diminishing omissions (Barat et al., 2023; Elkassem and Smith, 2023; Momenaei et al., 2023; Morath et al., 2023; Zumsteg and Junn, 2023), but medical information contextualization and abstract reasoning capability are needed. ChatGPT provides quite appropriate and readable medical knowledge with respect to diagnostic methods, visual prognosis, and surgical and nonsurgical treatments in terms of health-related counseling and factual information for both patients and physicians.

   ChatGPT produce factually precise and contextually significant structured discussion feedback to elaborate and emerging clinical questions, swiftly retrieving and integrating complex medical information into streamlined answers (Giannos and Delardas, 2023; Li et al., 2023; Rawashdeh et al., 2023), typifying coherent and safe medical practice criteria and patient care. ChatGPT can harness image analysis tools adequately, interpret clinical cases precisely, and evaluate treatment progress, resulting in enhanced personalized care.

   Users may harness ChatGPT to seek diagnostic information effortlessly without requiring physical appointment or medical expenses, and thus generative artificial intelligence technologies are pivotal in healthcare service limited access...