Here's How Big Data Analytics is Changing the Face of Precision Medicine "Big health data" generated from patient health records, diagnosis, genomic sequencing, medical research, smart devices, wearables holds the potential to transform healthcare & deliver personalized medicine

Precision or personalized medicine is a novel approach that considers a person's genetics, lifestyle, and environment to determine the best course of treatment and prevention. The short-term goal of precision medicine is to integrate and expand in oncology: understand genetic markers and find novel, more effective treatments. The long-term goal of precision medicine is to improve therapeutic tools in other fields of medicine.

Related: 4 Ways Savvy Entrepreneurs Stay Focused on Health When Traveling for Business

"Big health data" is the immense data generated from patient health records, diagnosis, treatments, genomic sequencing, medical research, smart devices, wearable, and various other sources. This data is available in high volume, can be transferred at high velocity, and is highly variable (since it is generated from multiple sources). As big health data grows larger, data analytics and data science have emerged as promising tools to harness its power. This data holds the potential to transform the healthcare industry, from precision medicine and drug research to population screening and effective decision making.

Related: Big Data Has the Potential to Transform Health Care

Key Steps in Precision Medicine Ecosystem: Generate Information, Transform Data, Deliver Insights

Multiple factors affect patients' health outcomes. These factors change through the course of treatment. A single/static diagnostic evaluation can limit clinicians' ability to precise prognosis and alter treatments based on resistance or response. Longitudinal patient health records now enable tracking of real-time clinical data of individual patients. These records capture information from clinical notes, lab reports, radiology imaging and scans, cells, tissues, fluids, organs, bio markers, inherited and acquired factors, individual lifestyles, environment, local/regional/subgroup population, and other assessments. Successful integration of these multiple data sets is key to precision medicine.

Sophisticated computational methods, algorithms, and advanced analytics are applied to convert heterogeneous data sets into clinically useful information. Machine Learning, AI, Natural Language Processing and Deep Learning enable faster and better mining of clinical data. Latest technologies are also being used to structured clinical data, identify patterns in the patient population, cluster/segregate population using clinical and molecular data points, identify mutations in large sets of genetic data, automate workflows, and help clinicians make informed decisions in real-time.

This enormous health data also requires secure and ethical handling. Block chain technology can be used to ensure that the data is shared and stored securely across different platforms.

All stakeholders in precision medicine ecosystem can utilize this data: healthcare providers can improve patient care, patients can educate themselves and collaborate with their physicians, researchers can improve study designs, health technology community can design and improve the data platforms, decision-makers can use this to maximize the health benefits and improve healthcare coverage and access.

Challenges with Utilization of EHRs and Genomic Data in India

Patient health records in the form of EHRs are currently heterogeneous – stored in multiple sources, formats, and databases. Different stakeholders use different semantics, vocabularies, terminologies, coding and classification systems for EHR data. Integrating and analysing this decentralized health data can thus be challenging. Health records can also be incomplete, inaccurate or inconsistent.

Insufficient amount of genomic data is collected in India due to limited research and development; India has around 18% of the world's population, but only represents less than 1% of existing global genetic data. Most research studies, drugs and treatments are based on Caucasian population and studies on Indian population are sparse or of low quality. Several startups in India are now looking to accelerate research in this area and enable the successful application of data from the Indian population.

Government health organizations are currently working on guidelines to ensure reliable and efficient identification, retrieval, access, and utilization of big health data. Ministry of Health and Family Welfare plans to digitize medical records in India. It recently published a National Digital Health Blueprint Report in the public domain to invite feedback from various stakeholders to streamline the adoption of digital technologies in health. It also released EHR Standards in 2016 to introduce a uniform system for maintenance of EHRs in the country.

India is also set to launch its first human genome mapping project in October this year. This project has been initiated by the Department of Biotechnology (DBT) and is called Genome India. It plans to catalogue genomic data of 10,000 Indians in its first phase.

Asia-Pacific Market for Precision Medicine

Asia-Pacific (APAC) region is the fastest-growing precision medicine market worldwide. An ageing population and the prevalence of chronic diseases have increased the healthcare burden in APAC. Countries like China and Japan are increasingly investing in genomics and healthcare technologies. Rising disposable income and economic growth will further boost this market.

According to a report by MarketWatch, APAC precision medicine market is expected to reach USD 21 Bn by 2023. It is projected to grow at a compound annual growth rate of 16.6%. This market is based on three segments:

• Ecosystem: pharmaceutical and biotechnology companies, clinical labs and diagnostic companies.
• Therapeutics: oncology, cardiovascular, psychiatric disorder, central nervous system, and infectious diseases.
• Technology: big data analytics, bioinformatics, gene sequencing, pharmacogenomics, and companion diagnostics.

Conclusions

To better diagnose, treat, and prevent diseases, it is important to understand the relationships between individual components of a larger system, where the whole is more than the sum of the individual components. These interrelations are often not visible when these components are individually investigated. The goal is to leverage technology to enable learning from each patient as more data is generated.

Medicine is all set to undergo a digital transformation - the future will see how healthcare community progresses from anatomy-based medicine to an all-inclusive, data-driven medicine that integrates data science, data analytics, digital health, and precision medicine. As new tools and infrastructure develop, collaboration and interoperability are increasingly important to maximize the potential of big health data.

Although this novel, data-driven approach may seem to challenge the conventional approach of hypothesis-driven medicine, its real promise lies in synergy and not a replacement. Both these approaches can be combined to improve clinical practice.