The concept of big data, which has risen manifolds in recent years, is going to be a turning point for the diagnostics sector
Worldover, diagnostics has been instrumental in generating huge data that can be analysed for clinical research and drug discoveries. As the sector continues to increase its focus on generating genetics and genetic mutation data, protein therapeutics data and designing personalised medicines, more and more of this information will be harvested to improve daily healthcare processes and solve complex health problems. Therefore, in the future, diagnostics is poised to become the biggest source for big data analystics.
In this article, we examine how big data has been utilised in the Indian diagnostic sector and what impact can be expected in the future.
The role of big data in Indian diangostics
In India, the diagnostics sector sees immense promise in big data to fuel growth for industry and gain actionable insights for clinical use. Experts believe that big data will be extensively utilised in the field of molecular diagnostics and genomic testing, finding new pathways to treat complex medical conditions and in furthering new drug discoveries in India.
Opines Dr Ravi Gaur, COO, Oncquest Labs, “Whenever we visit a doctor, generally he/she prescribes a series of diagnostic tests to figure out what is going on. Also, many times we undergo those tests without even getting conclusive results. Given that medical diagnostics is not an exact science, and often a diagnostic test is one result in a list of many such similar tests, the medical field is turning to big data for help. The concept of big data is based on the premise of mining a large database of similar data and comparing trends in that data to a case at hand. This can often assist the analysis of one patient’s case. Big data has been used for numerous studies but rarely for diagnostics. Of late, innovative thinking in the signal processing and data mining field aims to change that. If medical diagnostics could achieve that same efficiency using big data, the treatment of a variety of illnesses and maladies would change dramatically.”
According to Sanjeev Vashishta, MD and CEO, Pathkind Diagnostics, big data has huge relevance in diagnostics. It has changed the way we manage, analyse and leverage patients’ data to help them monitor their health, predict untoward incidents related to health and also to improve the quality of life. Huge volume and variety of data cannot be handled by conventional methods. Data mining at times has proven to predict diagnostics in a better way. It reduces clinical and economic burden of healthcare. Various studies highlight that more than 400,000 applications are available to facilitate self-health monitoring and the number of apps are going only northwards. Needless to mention, these applications have to bank on big data to churn out useful information.
Vashishta further explains that the ability to transform big data into actionable knowledge will disrupt at least three areas of medicine-Firstly, machine learning will drastically improve the ability of healthcare professionals to establish a diagnosis and make prognostics predictions based on thousands of risks predictive variable rather than at present prognostic modals based on a handful of variables like Acute Physiology and Chronic Health Evaluation (APACHE) score and Sequential Organ Failure Assessment (SOFA) score in the Intensive Care Unit. Better prediction would transform advanced care planning including optimum utilisation of costly ICU stay.
Secondly, machine learning would also display much of the routine work that a radiologist or Histo-pathologist does at present, as they focus on analysing patterns. Algorithms would make pattern recognition more accurate and faster besides being available 24*7 available rather than an eight hour shift and being as vigilant at 3 am as at 11 am and being available in here to inaccessible areas. Thirdly, algorithms will soon generate differential diagnostics, suggest high value test and reduce the over use of unnecessary tests. Further, experts speak about various segments of diagnostics that have potential to propel growth of big data.
Molecular diagnostics and genetics: A great source
Dr Vaidehi Jobanputra, Chairperson, Advanced Genomics Institute and Laboratory Medicine says, “Big data is a core part of molecular and genomic technologies because each assay generates a massive amount of data. Often, we are looking for changes and patterns in a huge volume of data which is akin to ‘searching a needle in a haystack’. So we need advanced computing and smarter approaches to derive meaningful and actionable insights. This is where the role of big data comes in.”
Experts also point out that big data is furthering scope for disease management and precision medicine as well. Hinting that big data analytics becomes a major source for disease management too, Dr Abhik Banerjee, Lab Director, Chief of Quality & Senior Consultant- Pathology, Suraksha Diagnostic adds, “Laboratory data should assist the clinician in providing better care for patients. In cases of insulin dependent diabetes mellitus (IDDM), retrospective review of huge lab data may potentially guide the physician to take decisions about appropriate doses of insulin which will ensure best intended effect. Similarly “big data” of laboratories will give more power to clinicians to choose appropriate doses of anticancer drugs for their patients with minimum side effects. Availability of big data with its rational management will definitely have a far reaching impact in the field of personalised medicine.”
More scope to precision medicine
The availability of patient data through multiple platforms has presented an opportunity to combine and consolidate data for devising personalised treatments. Says Amit Ray, Managing Director, Data Analytics, Protiviti Member Firm for India, “The emergence of value-based care has encouraged all healthcare stakeholders to work together towards eradicating certain high impact diseases such as diabetes and CVDs by using more and more data to drive better patient care at the lowest cost possible.”
According to Dr Banerjee, In the era of targeted therapy and precision medicine, availability of big data with laboratories will definitely give a momentum to researchers and scientists towards finding effective disease-specific and disease-modifying treatments. Hence big data will definitely be an integral part for research and innovation in therapeutic arenas in the near future.”
Dr Jobanputra emphasises, “Big data and genomics provide great opportunity and mechanism to look at each patient uniquely instead of broadly or forcibly categorising them to the nearest disorder or syndrome. Thus, seemingly unrelated clinical conditions and phenotypes can be studied and correlated with their genotypes. Such correlations open new avenues for precision medicine.”
The diagnostic industry has seen a shift from trial-and-error to targeted therapy for illnesses.
According to Dr Ravi Gupta, Chief Scientist Bioinformatics R&D, MedGenome Labs, clinicians will have significantly more time to spend treating patients and counselling them, than diagnosing a certain illness and coming to a conclusion.
“With the help of big data ,we can understand how a person’s genetics, environment, and lifestyle can help determine the best approach to prevent or treat disease. This has specifically helped in medical oncology. Increasing use of NGS in diagnostic centres has greatly simplified sequencing and decreased the costs for generating whole genome sequence data. The cost of complete genome sequencing has come down substantially. NGS technology has resulted in an increased volume of biomedical data that comes from genomic and transcriptomic studies. Combining the genomic and transcriptomic data with proteomic and metabolomic data can greatly enhance our knowledge about the individual profile of a patient—an approach often ascribed as “individual, personalised or precision healthcare” says Dr Gaur. Systematic and integrative analysis of omics data in conjugation with healthcare analytics can help design better treatment strategies towards precision and personalised medicine. This might turn out to be a game-changer in future medicine and health.”
Adds Vashishta, “Today, huge quantum of patient data is generated and using advanced predictive analyses, this information has the potential to enable diagnosis, treatment, and prevention of disease at a highly personalised level.”
Big business opportunity
Now from a business perspective also, the scope for big data in diagnostics is certainly wide. Many multinational tech gaints have already developing technologies backed by big data to tap the market and further invest in research.
Vashista informs that Google has established a number of new businesses that combine its big data capabilities with medical applications. The Google X research lab part of Google X Life Sciences is creating the baseline study with the goal to build a genomics database to facilitate early diagnosis and disease prevention. Google anticipates that this may become the world’s biggest data base, to include all sorts of medical information and laboratory test data.
Major IVD companies such as Beckman Coulter Diagnostics, Qiagen, Roche, Illumina, and PerkinElmer are gung ho about big data. “Data is being shared and gene targets have been validated and included in either targeted tests or panels. Now laboratory professionals and clinicians need tools to make sense of the continuous stream of tests and targets. IVD companies in collaboration with established and new IT companies are responding to this need with a massive wave of IT tool acquisitions, alliances and collaborations. These IT tools are offered as standalone services and also integrated into test platforms”, informs Vashishta.
Roadblocks to future growth
While industry leaders see a lot of scope for big data utilisation in diagnsotics, the biggest challenge at hand is integration and the lack of understanding among providers. The other challenges lie in the interpretation of astronomical data in a rapid, cost-effective and meaningful manner. Unfortunately, all the data used to construct the image is not necessarily helpful in a diagnosis, reveal experts. A lot of it is useless to the task of diagnosing the patient’s condition. One of the major obstacles is that a large amount of data remains stored in various de-centralised data servers that are inaccessible to researchers due to data protection laws.
Dr Gaur mentions, “Heterogeneity of data is a big challenge in data analysis. The huge size and highly heterogeneous nature of big data in healthcare renders it relatively less informative using conventional technologies. Advanced algorithms are required to implement machine learning and artificial intelligence approach for data analysis. A good knowledge of biology and IT is required to handle big data from biomedical research. Methods for big data management and analysis are being continuously developed. Storage, accuracy, correctness, consistency, relevancy, unified formats and security are some of the growing challenges. New technologies like blockchain can probably help in addressing cyber security issues and making patient data more secure.”
Ray mentions, “The biggest challenge to the optimal use of data is the lack of integration between different systems that produce or compile data in a healthcare system. This usually produces different sets of data for the same patient and discrepancy is usually found.”
Despite considerable advancements on the data analytics side, there are huge challenges, says Vashishta. Nevertheless, these roadblocks came be overcomed with smarter technologies and right strategies. Vashishta further explains, “In the Indian context, we still don’t have a common repository for patient data. In view of the fact that healthcare is still managed substantially by the unorganised sector, not many service providers are able to maintain patient records electronically. As a first step, we need to start registering patient data and then gradually link the same with a centralised repository system, access to which should be given to the service providers in the private, philanthropic and government space. This would help in better outcomes.”
Dr Gupta adds, “While migrating from physical to a cloud infrastructure, it is crucial to use the right tools to reduce security breaches. Limiting network access and controlling physical access and developing strategies to ensure safety can help protect sensitive patient data. However, there is a need for more enhanced patient data security and privacy guidelines in the Indian sub-continent as well.”
Dr Banerjee, warns, “Unauthorised access to patient data might be detrimental. Unique ID, password-enabled LIMS, certified cloud based applications, document control, face recognition or finger imprint capture facility of user are some of the security measures used by reputed laboratories in and outside India.”
What the future beholds?
In future, digital transformation holds the key to success. Well defined algorithms will be driven by scientific data produced by the diagnsotics sector. Artificial intelligence and augmented intelligence will help in arriving at finer and precise diagnosis. Therefore, big data analytics will continue to aid in therapy monitoring, disease prognosis, hereditary risk assessment and disease surveillance.
Dr Gupta mentions, “We feel that there are enough opportunities for diagnostic players to collaborate with the government and private companies on causes that will ease the burden on the healthcare ecosystem of the country. Aggregating, integrating patient data and applying big data analytical techniques will help in developing better models which will enable the doctors in making better clinical decisions. Further, analysing big data from patients will enable diagnostics service providers in developing newer solutions including developing screening programmes for various diseases that will benefit the masses.”
Dr Jobanputra states, “Diagnostic organisations need to define their area of focus and expertise. A lot of technology is available off the shelf, but technology should not be confused with expertise.” Dr Vashishta sums up, “Big data could play a pivotal role in building scenarios that would help standardising protocols to handle some of the basic protocols and also help paramedics to take informed calls in remote areas where there is dearth of doctors/ specialists.”
As rightly pointed out by experts, big data does hold the key to success for the diagnsotic sector in India. It has tremendous potential to improve healthcare. But the question of appropriate intregration still remains unanswered. The industry must work on educating healthcare providers on the proper intepretation and integration of big data. Also, the industry will need to address challenges associated with data privacy, security, ownership, and governance in future.