Artificial Intelligence in Healthcare

Technology applications apps encourage healthier  behaviour in individuals and help in the proactive management of healthy lifestyle. Additionally AI increases the ability for healthcare professionals to better understand to day to day patterns and needs of people they care for better feedback ,care and support .Health monitoring tools such as: Wearable health trackers like those from FITBIT, Apple ,Garmin and others monitors health rate and activity levels. They can send alerts to the users to get more exercise and can share this information to the doctors. AI is already use to detect diseases such as cancer more accurately in their early stages. According to American Cancer Society a high proportion of Mammograms yield false result leading to one in two healthy women being told they have cancer . The use of AI is enabling review and translation of mammograms 30 times faster with 99% accuracy reducing the need for unnecessary biopsies . Googles deep mind health is working in partnership with clinicians ,researchers ,and patients to solving detecting real world healthcare problems.AI can be useful in many ways like: Digital consultation ,Treatment design AI- robot assistant surgery,  epidemic  prediction disease surveillance .

                         MEDICAL IMAGING :- Machine learning algorithms can process unimaginable amounts of information in the blink of an eye and provide more precise than humans in spotting even the smallest detail in medical imaging .

Digital consultancy:- For example the digital health firms Health Tap developed Dr .A.I. and apps like  Babylon in the UK use AI to medical consultation based on personal medical history and common medical knowledge . Users report their symptoms to the apps which uses speech recognition to compare against the database of illness and ask patient to specify symptoms to triage whether they should go to the ED ,urgent care . 

Treatment Design :- AI system have been created to analyze data -notes and reports from the patients 's life ,external research and clinical expertise - to help select the correct individually customized treatment path.

Research on Molecular Epidemiology :- Recently, the greatest statistical computational challenges in molecular epidemiology is to identify and characterize the genes that interact with others genes and environmental factors that bring the effect on complex multifactorial disease.

                                       This phenomenon can be solved traditional statistical method due to high dimensionality of the data and occurrence of the multiple polymorphism . Hence, there are several machines learning methods to solve such problems by identifying such susceptibility gene which are neutral networks , SVM and RFs in such common multifactorial disease.

 REFERENCES :-AI IN INDIA PDF, //ESCAP AI PDF

Artificial Intelligence in Healthcare

Medical practice is being influenced by artificial intelligence (AI). AI applications are moving into domains that were previously regarded as only the domain of human expertise because of recent advances in digitized data collecting, machine learning, and computing infrastructure. ^[1]

Machine learning methods for structured data, such as the classic support vector machine and neural network, and current deep learning, as well as natural language processing for unstructured data, are all popular AI techniques. Cancer, neurology, and cardiology are three major illness areas that use AI techniques.

In this discipline, the IBM Watson system is a pioneer. The system, which contains both machine learning and natural language processing modules, has shown encouraging results in cancer. Watson's therapy recommendations, for example, are 99 per cent consistent with physician conclusions in a cancer study. 66 Watson also partnered with Quest Diagnostics to provide an AI Genetic Diagnostic Analysis. Furthermore, the system began to affect actual clinical practices. Watson, for example, successfully detected the rare secondary leukaemia caused by myelodysplastic syndromes in Japan by analyzing genetic data. ^[2]

Reference

1. Yu KH, Beam AL, Kohane IS. Artificial intelligence in healthcare. Nature biomedical engineering. 2018 Oct;2(10):719-31.

//www.nature.com/articles/s41551-018-0305-z

2. Jiang F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, Wang Y, Dong Q, Shen H, Wang Y. Artificial intelligence in healthcare: past, present and future. Stroke and vascular neurology. 2017 Dec 1;2(4).

//svn.bmj.com/content/svnbmj/2/4/230.full.pdf

John McCarthy in 2004 defined Artificial intelligence (AI) (1) as “the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.” AI has applications across diverse fields and integrating it with healthcare can bring about a paradigm shift. This paper presents a structured literature review of the AI in healthcare (2). The authors find AI to be a diverse field that attracts an increasing number of researchers with the United States, China, and the United Kingdom contributing to the highest number of studies. It further discusses the applications of AI in healthcare.

References

1. //homes.di.unimi.it/borghese/Teaching/AdvancedIntelligentSystems/Old/IntelligentSystems_2008_2009/Old/IntelligentSystems_2005_2006/Documents/Symbolic/04_McCarthy_whatisai.pdf
2. //bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-021-01488-9

In healthcare, the most common application of traditional machine learning is precision medicine – predicting what treatment protocols are likely to succeed on a patient based on various patient attributes and the treatment context.[4]

Efficiency in early detection:

According to a study by American Cancer Society, a greater number of mammograms yield leading to false detection of cancer in healthy women. Using AI can enable review and translation of mammograms 30 times faster with 99% accuracy, reducing the requirement for avoidable biopsies.[1]

Importance in diagnosis:

Google’s DeepMind Health has been working alongside scientists, physicians and patients to solve healthcare problems by combining machine learning and neuroscience to build powerful general-purpose learning algorithms into neural networks that mimic the human brain.

Contribution to treatments:

AI can help doctors take a more comprehensive approach for disease management, detecting deterioration of health due to lifestyle in patients, better coordinate care plans and help patients to better comply with long-term treatments.

Role in research:

According to the California Biomedical Research Association, it takes an average of 12 years for a drug to travel from the research lab to the patient.[3] Implementation of the latest advances in AI to facilitate drug discovery and processing can cut both the time to market new drugs and their costs.

Future concerns on displacement of human workforce by AI:

There has been considerable attention to the concern that AI will lead to automation of jobs and substantial displacement of the workforce. The limited incursion of AI into the industry thus far and the difficulty of integrating AI into clinical workflows has not yet affected human workforce. It thus seems likely that the healthcare jobs most likely to be automated would be those that involve dealing with digital information, radiology and pathology for example, rather than those with direct patient contact.

 There are also a variety of ethical implications has healthcare decisions have been made almost exclusively by humans in the past and the use of smart machines to make or assist with them raises issues of accountability, transparency, permission and privacy.

The greatest challenge to AI in healthcare domains is ensuring their adoption in daily clinical practice.[5]

REFERENCES:

1. //www.wired.co.uk/article/cancer-risk-ai-mammograms
2. //www.pwc.com/gx/en/industries/healthcare/publications/ai-robotics-new-health/transforming-healthcare.html
3. //www.ca-biomed.org/pdf/media-kit/fact-sheets/CBRADrugDevelop.pdf
4. A machine learning approach to integrate big data for precision medicine in acute myeloid leukemia. Lee SI, Celik S, Logsdon BA, Lundberg SM, Martins TJ, Oehler VG, Estey EH, Miller CP, Chien S, Dai J, Saxena A, Blau CA, Becker PS. //www.ncbi.nlm.nih.gov/pmc/articles/PMC5752671/
5. Davenport T, Kalakota R. The potential for artificial intelligence in healthcare. Future Healthc J. 2019;6(2):94-98. doi:10.7861/futurehosp.6-2-94 //www.ncbi.nlm.nih.gov/pmc/articles/PMC6616181/

AI has the potential to transform many aspects of patient care. The important applications involve diagnosis and treatment recommendations, patient engagement and adherence, and administrative activities. Patient engagement and adherence has long been seen as the final barrier between ineffective and good health outcomes. The more patients proactively participate in their own well-being and care, the better the outcomes. These factors are now increasingly being addressed by AI. Clinical experts develop a plan of care for their patients to improve their chronic or acute. However, that often doesn't matter if the patient fails to make the necessary behavioral adjustment, e.g. losing weight, scheduling a follow-up visit, or complying with a treatment plan. Noncompliance is a major problem. However if deeper involvement by patients results in better health outcomes, can AI-based capabilities be effective in personalizing and contextualizing care? There is growing emphasis on using machine learning and business rules engines to drive nuanced interventions along the care continuum. Messaging alerts and relevant, targeted content that provoke actions at moments that matter is a promising field in research.

References

//www.ncbi.nlm.nih.gov/pmc/articles/PMC6616181/

The study “Use of AI-based tools for healthcare purposes: a survey study from consumers’ perspectives” focuses on AI (artificial intelligence) based CDS (clinical decision support) features. The Study was conducted in USA by using online survey of 307 individuals.   It was found that technological (perceived performance anxiety and perceived communication barriers), ethical perceived privacy concerns, perceived mistrust in AI mechanism, and perceived social biases factors), and regulatory concerns (perceived unregulated standard and perceived liability issues) influence the risks belief of using AI applications in healthcare.  Out of these three, technological concerns was the most significant predictors of risk belief.  In order to make people use the AI tools it is important to address and analyze the technological, ethical, and regulatory concerns.

read here: //www.ncbi.nlm.nih.gov/pmc/articles/PMC7376886/

The healthcare industry is at the verge of receiving major changes. The advancement of technology in the healthcare field will not only deploy more precise, efficient, and impactful interventions for patient’s care but will also help in successfully assessing, diagnosing and treating chronic diseases, and cancer etc. The large amount of data related to patient care can be stored and used at the right time using artificial intelligence which will definitely help in improvement across the patient care continuum. AI offers a number of advantages over conventional analytics and clinical decision-making procedures. Interaction with data algorithms will allow humans to gain insights into variable diagnostics, care and treatment processes, and patient outcomes. Also, in imaging analysis, the rapid advances in AI will mostly aid in examining radiology and pathology images by a machine. Although, speech and text recognition are already employed for the purpose of patient communication and documenting clinical notes, leading to their increased usage.

However, implementation issues with AI strains out many healthcare organisations. Although rule-based systems incorporated within EHR systems are widely used but they lack precised algorithmic systems based on machine learning. These rule-based clinical decision support systems are difficult to maintain as medical domain keeps on updating but are often unable to handle the big data and knowledge based on multiple care approaches such as genomic, proteomic, metabolic and other ‘omic-based’.

References:

1. //healthitanalytics.com/news/top-12-ways-artificial-intelligence-will-impact-healthcare
2. //www.ncbi.nlm.nih.gov/pmc/articles/PMC6616181/

Pattern detection, prediction, classification, language processing, and image recognition are machine learning functions seen in consumer-oriented services, such as video recommender systems, chatbots, and facial recognition. Applications of such functions are seen in the area of nutrition as well. Some domains include diet optimization, food image recognition, risk prediction, and diet pattern analysis.
For Example, Noah et al. (2004) proposed an artificial intelligent menu planning system named DietPal. It is a web-based system for producing dietary and menu management. The DietPal was built for Malaysian dieticians and medical professionals of health centers. The system assisted them in creating healthy menus for patients based on their health problem history that defined menu constraints. The system benefits users in planning meals effectively with accurate nutrition requirements based on the user’s health history. Their study used trial and error exchange menu items approach with the food groups based on the guidelines from nutritionists. But, user preference was not taken into account to generate healthy meals which can lead to the dissatisfaction of users.

Many research opportunities are still available to study the use of Artificial Intelligence in various domains of healthcare.

Interesting Reads : 

1.Noah, S.A., S.N. Abdullah, S. Shahar, H. Abdul-Hamid and N. Khairudin et al., 2004. DietPal: A web-based dietary menu-generating and management system. J.Med. Internet Res., 6: e4-e4. DOI: 10.2196/jmir.6.1.e4 

2.Limketkai, B.N., Mauldin, K., Manitius, N. et al. The Age of Artificial Intelligence: Use of Digital Technology in Clinical Nutrition. Curr Surg Rep 9, 20 (2021). //doi.org/10.1007/s40137-021-00297-3