Healthcare and laboratory medicine continue to improve upon existing technologies and introduce new innovations to improve patient care. Indeed, healthcare technology allows for increased efficiency, reduces human errors, and can also enhance patient satisfaction. Historically, upgrades to health technology have been relatively slow and fragmented. It is very difficult to broadly execute technological innovations within healthcare in the United States because of the fractured health system. Interestingly, the novel SARS-COV-2 virus outbreak and the COVID-19 pandemic provided an opportunity for mass healthcare technology improvements. It is estimated 10 years of technological upgrades and improvements were implemented in just a few short months during the early part of the pandemic. While BeaconLBS has previously reviewed different types of healthcare technology, the purpose of this article is to discuss how artificial intelligence, machine learning, chatbots, and clinical decision support systems may improve healthcare and patient outcomes.
Artificial Intelligence (AI) is a broad term that describes the concept of computer software that can replicate, copy, or improve decision making without human intervention. There are many different types of AI technologies, with different strengths and weaknesses. Machine Learning (ML) is a type of AI that is trained to recognize patterns using statistics and mathematical modeling to make decisions. For example, ML has been applied to dermatology to detect skin cancer. The process of evaluating a patient for skin cancer can be cumbersome, which may include a visual inspection by a dermatologist, digital photographs, and a biopsy sent to a lab for microscopic analysis. Skin biopsies can be evaluated based on color, border, asymmetry, diameter, and progression over time. Additionally, early detection allows for better health outcomes. Multiple ML technologies have been trained and validated to detect skin cancer. One research group was able to achieve 94% sensitivity and 97% specificity in evaluating skin cancer with ML. Digital visualization ML has also been applied to the emerging field of digital pathology.
A chatbot is an AI computer program that mimics human language and conversation while interfacing with patients. Chatbots may increase patient engagement and improve medical adherence. Furthermore, chatbots are scalable and cost-effective. In response to the COVID-19 pandemic, the World Health Organization (WHO) Regional Office for Europe launched a health app called HealthBuddy+ in collaboration with local health authorities and other partners. HealthBuddy+ became a communication tool between healthcare officials and the general population. The chatbot allowed for the agile dissemination of new information as data from the pandemic was quite fluid at times. The chatbot expanded from a web-based tool at launch to a fully functional mobile health app available in 16 regional languages. Chatbots also have clinical utility in hereditary cancer, mental wellness, and other medical specialties.
Clinical Decision Support Systems (CDSS) may improve healthcare by assisting physicians with medical management decisions and can be integrated into the Electronic Health Record (EHR). CDSS rules may be built on literature, practice guidelines, or clinical evidence. When incorporated into the EHR, CDSS may warn a clinician of an adverse medication interaction or can reduce redundant laboratory testing with a “popup.” CDSS can be based on different AI, ML, or other mathematical modeling systems to generate the appropriate output. There is a wealth of literature published on the clinical utility of CDSS in various specialties including diagnostics, disease management, and medication management, among others.
Moreover, CDSS may be integrated into the EHR to assist with ordering lab tests and preauthorization. BeaconLBS can incorporate our proprietary laboratory management technology with 14 EHRs and 49 labs in the United States. Thus, the prior authorization process becomes more convenient and streamlined for providers, since prior authorizations are incorporated into the standard electronic test ordering process. Furthermore, it takes only seconds, instead of minutes, which helps physicians spend less time doing administrative duties associated with patient care.
However, as with any new technology that is rapidly evolving, there are challenges and ethical hurdles to consider. Privacy is of the utmost importance to many patients and providers. Medical records are considered Personal Health Information (PHI). The Health Insurance Portability and Accountability Act (HIPAA) was put in place to give patients more control of who can legally review their medical records. Additionally, there must be sufficient encryption and safeguards to prevent a data breach. Nevertheless, regardless of any encryption, the risk of a data breach of information that is stored electronically on servers, online, or in the cloud is never zero. Quality control is another issue that must be addressed. How is consistency ensured across multiple health systems and platforms? What are the acceptable margins of error for these technologies? Are healthcare providers equipped to answer questions regarding the process and the reasons why AI makes a specific decision? Should these technologies be incorporated into medical school, since it is likely they will be a part of routine clinical care, regardless of specialty? The questions are many and answers can be challenging.
AI, ML, chatbots, and CDSS are being incrementally applied to patient care, patient education, laboratory medicine, and preauthorization. The COVID-19 pandemic presented an opportunity to accelerate the implementation of some of these technologies on a global scale. BeaconLBS utilizes CDSS to make lab testing and preauthorization more efficient and cost-effective. However, there are many risks being addressed including privacy, safety, and encryption. It will likely take patients, clinicians, the private sector, professional medical societies, and regulatory agencies working together to maximize the potential utility and safety of these different health technologies.
Azam AS, Miligy IM, Kimani PK, Maqbool H, Hewitt K, Rajpoot NM, Snead DRJ. Diagnostic concordance and discordance in digital pathology: a systematic review and meta-analysis. J Clin Pathol. 2021 Jul;74(7):448-455. PMID: 32934103.
Casal-Guisande M, Comesaña-Campos A, Dutra I, Cerqueiro-Pequeño J, Bouza-Rodríguez JB. Design and Development of an Intelligent Clinical Decision Support System Applied to the Evaluation of Breast Cancer Risk. J Pers Med. 2022 Jan 27;12(2):169. PMID: 35207657.
Davenport T, Kalakota R. The potential for artificial intelligence in healthcare. Future Healthc J. 2019 Jun;6(2):94-98. PMID: 31363513
Li J, Yuan Y, Bian L, Lin Q, Yang H, Ma L, Xin L, Li F, Zhang S, Wang T, Liu Y, Jiang Z. A comparison between clinical decision support system and clinicians in breast cancer. Heliyon. 2023 May 5;9(5):e16059. PMID: 37215843.
Lin S, Nateqi J, Weingartner-Ortner R, Gruarin S, Marling H, Pilgram V, Lagler FB, Aigner E, Martin AG. An artificial intelligence-based approach for identifying rare disease patients using retrospective electronic health records applied for Pompe disease. Front Neurol. 2023 Apr 21;14:1108222. PMID: 37153672.
Melarkode N, Srinivasan K, Qaisar SM, Plawiak P. AI-Powered Diagnosis of Skin Cancer: A Contemporary Review, Open Challenges and Future Research Directions. Cancers (Basel). 2023 Feb 13;15(4):1183. PMID: 36831525.
Nazareth S, Hayward L, Simmons E, Snir M, Hatchell KE, Rojahn S, Slotnick RN, Nussbaum RL. Hereditary Cancer Risk Using a Genetic Chatbot Before Routine Care Visits. Obstet Gynecol. 2021 Dec 1;138(6):860-870. PMID: 34735417.
Rambaud K, van Woerden S, Salvi C, Smallwood C, Rockenschaub G, Faruqui N, Melo Bianco V, Mosquera M. Building a chatbot in a pandemic. J Med Internet Res. 2023 Mar 16. doi: 10.2196/42960. PMID: 37074958.
Sadasivan C, Cruz C, Dolgoy N, Hyde A, Campbell S, McNeely M, Stroulia E, Tandon P. Examining Patient Engagement in Chatbot Development Approaches for Healthy Lifestyle and Mental Wellness Interventions: Scoping Review. J Particip Med. 2023 May 22;15:e45772. PMID: 37213199.
Siglen E, Vetti HH, Lunde ABF, Hatlebrekke TA, Strømsvik N, Hamang A, Hovland ST, Rettberg JW, Steen VM, Bjorvatn C. Ask Rosa – The making of a digital genetic conversation tool, a chatbot, about hereditary breast and ovarian cancer. Patient Educ Couns. 2022 Jun;105(6):1488-1494. PMID: 34649750.
Sloane EB, and Silva RJ, Chapter 83 – Artificial intelligence in medical devices and clinical decision support systems, Editor(s): Ernesto Iadanza, Clinical Engineering Handbook (Second Edition), Academic Press, 2020, Pages 556-568.
Sutton RT, Pincock D, Baumgart DC, Sadowski DC, Fedorak RN, Kroeker KI. An overview of clinical decision support systems: benefits, risks, and strategies for success. NPJ Digit Med. 2020 Feb 6;3:17. PMID: 32047862.
Ting Sim JZ, Fong QW, Huang W, Tan CH. Machine learning in medicine: what clinicians should know. Singapore Med J. 2023 Feb;64(2):91-97. PMID: 34005847.
Zia Ur Rehman M, Ahmed F, Alsuhibany SA, Jamal SS, Zulfiqar Ali M, Ahmad J. Classification of Skin Cancer Lesions Using Explainable Deep Learning. Sensors (Basel). 2022 Sep 13;22(18):6915. PMID: 36146271.
HealthBuddy+: Access to trusted information on COVID-19 in local languages using an interactive web- and mobile-based application. World Health Organization. Accessed 06/24/2023. https://www.who.int/news-room/feature-stories/detail/scicom-compilation-healthbuddy
Healthcare has made 10 years of progress in just a few months. Here’s how. LinkedIn. Accessed 06/24/2023.
Summary of the HIPAA Privacy Rule. U.S. Department of Health and Human Services. Accessed 06/24/2023. https://www.hhs.gov/hipaa/for-professionals/privacy/laws-regulations/index.html