Dr Srivatsa P, AGM, Group Technical Co-ordinator, Apollo Diagnostics, Hyderabad reviews the impact of some advances in diagnostics like AI, internet of medical things, predictive analytics and the omics, and opines that the way of the future is to provide quick and quality reports to end users such as customers, patients and clinicians with improved user-friendly accessibility
The market for diagnostic services has been growing in India over the past couple of years at a rate of 15-20 per cent and is expected to grow at a CAGR of approximately 16 per cent accounting to approximately $802 billion in the financial year 2020. Within the diagnostics market, the pathology segment is estimated to contribute approximately 58 per cent of total market, by revenue. The current rate of growth is expected to continue, driven by increasing awareness in the population, improving payer coverage and rising incidence of lifestyle diseases.
Artificial intelligence in health care
COVID-19 has greatly accelerated the use of telehealth resources. In April 2020, 43.5 per cent of Medicare primary care visits utilised telehealth methods rather than in-person visits. One of the major benefits of telehealth over in-person alternatives is that it reduces contact between patients, healthcare workers, and other patients. Wearable devices enable healthcare workers to have real-time information on patient data while they remain at home.
More importantly, telehealth’s growth appears likely to continue even after the pandemic is over. This boom in telehealth seems likely to break $185.6 billion by 2026.
With the advent of diagnostic services algorithms can provide aids in diagnosing complex situations. AI is poised to become a transformational force in healthcare by combined inputs from machine learning. AI helps in developing next generation tools to drill down to the pixel level of imaging thereby providing clearer data to analyse. In other applications such as neurology and radiology, AI is being used to enhance neural network simulation to make informed decisions faster to save lives and provide better treatment. Smart monitoring devices attached to patients can monitor, scan and evaluate data and provide suggestions to clinicians on a faster note regarding impending sepsis, mental deterioration, possibilities of a stroke and also immunotherapy options by analysing responses to cancer treatment. Electronic health records are fool proofed for data integrity and quality issues by negating biases. By powering a new generation of tools and systems that make clinicians aware in decision-making processes, AI shall transform the medical field to the better for patient care. The major concern of data privacy and regulatory compliance from the social angle is debatable in all affairs concerning the use of data science in healthcare.
Internet of medical things
By 2025, the IoT industry will be worth $6.2 trillion. The healthcare industry has become so reliant on IoT technology in 2020 that 30 per cent of that market share for IoT devices will come from healthcare. Almost all consumers have access to devices with sensors to collect healthcare data. Smartphones, tablets, wearable devices, collect data from an individual and submit for analysis for appropriate actions to be taken for the individual’s benefit. Big data analytics help in storing and statistically analysing the data for informed decisions to be taken by a medical professional. Ongoing research also suggests that ingested nanoparticles can gather information on occult cancers, impending cardiac plaques and critically low blood levels of compounds and submit the same to healthcare experts from one patient, so treatment can begin at the earliest.
The global predictive analytics in healthcare market was valued at $1,806 million in 2017, and is estimated to reach $8,464 million at a CAGR of 21.2 per cent from 2018 to 2025. It helps doctors and healthcare workers make treatment related decisions based on current data available regarding a consumer’s health. It describes a methodology wherein insights are possible into future events to answer questions such as “What might happen in this case?” This entertains tailor made treatment strategies thereby shortening time taken for cure to occur. Predictive analytics finds a pattern in historical and transactional data and uses it to identify risks and opportunities for the future. Based on the available descriptive data, predictive analytics uses different techniques, which include machine learning, statistical techniques, and predictive modelling to evaluate and determine the probable future. The purpose of predictive algorithms in healthcare is:
- To find the correlations in the patient’s data;
- To find associations of the symptoms;
- To find familiar antecedents of the symptoms;
- To explore the impact of different factors (genome structure, clinical variable, et al.) on the course of treatment;
- To examine the possible influence of past and current diseases.
Depending on the goal of the analysis, a predictive algorithm can produce assumptions based either on available data directly from a given patient or general medical data from the public health datasets. It’s important to remember that predictions are, in fact, nothing more than assumptions and probabilities.
Genomics, proteomics and metabolomics are those branches of healthcare which create opportunities for personalised medicine. Genomics’ based gene chips caries numerous nucleic acids on a microarray allowing the possibility of many genes to be sequenced to figure out genetic abnormalities in patients with cancer, autoimmune disorders, cystic fibrosis and congenital abnormalities. Microfluidic systems enable QRT-PCR based sequencing that measure RNA levels related to gene transcription processes. Proteomics deals with proteins expressed by abnormal genes in relation to gene expression. Proteomes characterised in blood and body fluids characterise post translational modifications. Protein profiling identifies thousands of proteins through novel biochemical mass spectrometric and laser assisted detection methods. These serve as diagnostic or prognostic markers in diseases related to new born especially. Metabolomics is the systematic study of chemical metabolites fingerprinting chemical processes by means of study of small molecules. It is an analytical profiling technique measuring large amounts of metabolites in blood and body fluids. Combining high throughput analytical chemistry and data analysis with viruses because they rewire host metabolism, novel biomarker discovery is easier. Mass spectrometry based techniques such as GCMS quantifies cellular metabolites from vitamins and amino acids, xenobiotic compounds from the diet or environment. Physiochemical properties are analysed and reported from samples to detect abnormal levels of minor metabolites otherwise undetected in blood causing anomalies in physiology of patients.
Healthcare sector in India continues to see a sharp growth driven by increasing incidence of lifestyle diseases and improvements in technology and procedures. The way of the future is to provide quick and quality reports to end users such as customers, patients and clinicians with improved user-friendly accessibility.