Next Healthcare: When Big Data became the new Normal

Drug Discovery Enabled by Big Data, FHIR, Next-Generation Sequencing, and Artificial Intelligence

The healthcare industry is going through a major shake-up as big data becomes the new normal. Thanks to the adoption of Fast Healthcare Interoperability Resources (FHIR) standards and the advancements in next-generation sequencing (NGS) of genomes, this transformation is well underway. And now, there's an exciting addition to this revolution: AI-enabled precision medicine. In this article, we'll explore how these innovations are reshaping healthcare and paving the way for more personalized and effective treatments.

Interoperability and FHIR in Healthcare

The healthcare industry has always relied heavily on data, and the shift towards big data has had a profound impact. Health Level Seven International's (HL7) FHIR standard has revolutionized healthcare data management by providing a standardized framework for the electronic exchange of healthcare information. This ensures the interoperability of electronic health records across different healthcare systems (1). FHIR has transformed the way healthcare data is managed, offering a standardized, easy-to-implement, and rigorous approach for exchanging data between healthcare applications. The adoption of FHIR has opened the door to more patient-centered and integrated care (1).

Machine-Accessibility 

Fast Healthcare Interoperability Resources (FHIR) has emerged as a crucial standard in healthcare data administration, offering a standardized, easy-to-implement, and rigorous method for exchanging data between healthcare applications. The adoption of FHIR has paved the way for more integrated and patient-centered care by closely aligning with the FAIR (Findable, Accessible, Interoperable, and Reusable) principles, which provide guidelines for improving the management of scientific data (1). The FAIR principles highlight the importance of machine-actionability, allowing computational systems to easily locate, access, interoperate, and reuse data with minimal human intervention (5). This is particularly relevant in the healthcare context, where the amount, complexity, and speed of data creation have significantly increased.

Findability

FHIR adheres to the 'Findable' principle of FAIR by assigning globally unique and persistent identifiers to data and metadata, making them easily discoverable by both humans and computers (1), (5). FHIR guarantees that data and metadata can be retrieved through their identifier using a standardized communications protocol (1), (5), thereby fulfilling the "accessible" principle.

Interoperability

When it comes to interoperability, FHIR utilizes a formal, accessible, shared, and widely applicable language for representing knowledge. Additionally, it incorporates FAIR-compliant vocabularies and includes qualified references to other data and metadata (1), (5).

Reusability

FHIR's approach to data administration incorporates the principle of "reusability." FHIR ensures that data and metadata are thoroughly described with a wide range of accurate and relevant attributes. Additionally, they are released with a clear and accessible data usage license, associated with a detailed provenance, and conform to community standards relevant to the domain (1), (5).

In summary, FHIR plays a crucial role in promoting data interoperability in healthcare and is closely aligned with the FAIR principles. As a result, it serves as a vital contributor to the transition towards big data in healthcare.

The Beginning of Personalized Medicine

The advent of next-generation sequencing (NGS) has completely transformed the field of genomics. It allows for the rapid sequencing of large portions of an individual's genome, enabling personalized medicine and significantly advancing our understanding of genetic diseases (2). Moreover, the cost of sequencing the human genome has significantly decreased over time, making it more accessible for both research and clinical applications (4).

NGS empowers us to comprehensively analyze a patient's genome, providing valuable insights into their genetic predispositions to specific diseases, drug responses, and other health-related characteristics. This wealth of information can be utilized to personalize disease prevention, diagnosis, and treatment (6), tailoring medical interventions to the unique genetic makeup of each individual.

However, the potential of NGS in personalized medicine does come with its fair share of challenges. One major obstacle is the interpretation of the vast amount of data generated by NGS. This involves identifying clinically relevant variants among the millions of variants found in an individual's genome. To address this issue, guidelines and standards for the interpretation of genetic variants have been devised by the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) (7).

Another obstacle is the integration of genomic data with other types of data, such as electronic health records (EHRs). While integrating EHRs with genomic data can provide a more comprehensive view of a patient's health, it requires robust data standards and interoperability to ensure effective sharing and utilization of data across different platforms and healthcare settings (6).

Fortunately, machine learning and artificial intelligence (AI) can enhance the potential of NGS in personalized medicine. These technologies have the ability to analyze large volumes of data, identify patterns, and make predictions, which can be used to inform personalized healthcare decisions (8).

The Future of Healthcare in P4

Precision medicine, also referred to as personalized medicine, stratified medicine, or P4 medicine (predictive, preventative, personalized, and participatory), is an innovative approach to healthcare that customizes disease prevention and treatment based on the genetic variability of each individual. It encompasses a comprehensive strategy that includes extensive biomarker testing, close monitoring, in-depth statistical analysis, and health counseling for patients. The ultimate goal is to detect and prevent diseases at the earliest possible stage, thereby enhancing patient outcomes and reducing healthcare costs.

Pharmacogenomics, an intriguing field where precision medicine is making a significant impact, explores how genes influence an individual's response to medications. This field combines pharmacology (the study of medicines) and genomics (the study of gene functions) to develop medications and dosages that are effective, safe, and tailored to a person's genetic makeup.

The Function of Medical Records and Standards

Health records, especially electronic health records (EHRs), are crucial to precision medicine. They contain a plethora of information regarding the medical history of a patient, including diagnoses, treatments, and outcomes. This information can be analyzed to identify patterns and correlations that can inform individualized treatment plans.

However, the full potential of EHRs in precision medicine cannot be realized unless their data is standardized. This is where standards for health data such as FHIR come into play. FHIR (Fast Healthcare Interoperability Resources) is a standard for the electronic exchange of health information. It facilitates the integration and interoperability of health data, making it simpler for healthcare providers to access and share patient data.

FHIR can facilitate the integration of genomic data into EHRs for precision medicine. This enables healthcare providers to access a patient's genomic information alongside their medical history, allowing them to make more informed treatment decisions.

Surmounting Obstacles Using AI and Machine Learning

Despite the promise of precision medicine, its widespread implementation faces significant obstacles. The sheer volume and complexity of health data, which can be difficult to manage and analyze, is one of the greatest obstacles.

Artificial intelligence (AI) and machine learning (ML) enter into play here. Large volumes of health data can be analyzed by AI and ML algorithms, which can then identify patterns and make predictions. They can assist physicians in making sense of complex genomic data and identifying the most effective treatments for individual patients.

A study published in the Journal of the American Medical Informatics Association, for instance, demonstrated the use of AI and ML for clinical phenotyping using EHR data. The study discovered that AI technologies applied to unstructured EHR data outperformed traditional query techniques on structured EHR data for each evaluated concept, indicating that advanced methodologies utilizing unstructured data are required to enhance performance impact (9).

Conclusion

Precision medicine is the future of healthcare, promising treatments that are as unique as a unicorn's horn. With the integration and interoperability of health data, thanks to health records and standards like FHIR, we're taking giant leaps towards a healthcare utopia. And let's not forget about the superheroes AI and ML, swooping in to analyze complex health data and pave the way for precision medicine, even in the face of obstacles.

The rise of big data in healthcare, fueled by FHIR standards, next-generation sequencing, and AI-enabled drug discovery, signals the dawn of a new era in healthcare delivery and research. As we ride the wave of big data and AI, we can expect healthcare solutions that are tailor-made, efficient, and mind-blowingly effective. But, hold on tight! This also means we need to step up our game in terms of data governance and ethical considerations. We've got a responsibility to ensure that big data and AI are used for the greater good in healthcare. Let's make it happen!

References

1. http://hl7.org/fhir. Overview - FHIR v5.0.0. Overview - FHIR v5.0.0 https://www.hl7.org/fhir/overview.html .
2. Dash, S., Shakyawar, S. K., Sharma, M. & Kaushik, S. Big data in healthcare: management, analysis and future prospects - Journal of Big Data. Springer Open https://journalofbigdata.springeropen.com/articles/10.1186/s40537-019-0217-0 (2019) doi:10.1186/s40537-019-0217-0  .
3. Kulski, J. Next-Generation Sequencing — An Overview of the History, Tools, and “Omic” Applications. the UWA Profiles and Research Repository https://research-repository.uwa.edu.au/en/publications/next-generation-sequencing-an-overview-of-the-history-tools-and-o-2 (2017) doi:10.5772/61964 .
4. The Cost of Sequencing a Human Genome. Genome.gov https://www.genome.gov/about-genomics/fact-sheets/Sequencing-Human-Genome-cost (2021) .
5. FAIR Principles - GO FAIR. GO FAIR https://www.go-fair.org/fair-principles/ .
6. Personalized Medicine and the Power of Electronic Health Records. Personalized Medicine and the Power of Electronic Health Records - ScienceDirect https://www.sciencedirect.com/science/article/pii/S0092867419302223 (2019) doi:10.1016/j.cell.2019.02.039 .
7. InterVar: Clinical Interpretation of Genetic Variants by the 2015 ACMG-AMP Guidelines. InterVar: Clinical Interpretation of Genetic Variants by the 2015 ACMG-AMP Guidelines - ScienceDirect https://www.sciencedirect.com/science/article/pii/S0002929717300046 (2017) doi:10.1016/j.ajhg.2017.01.004 .
8. Radiomics: Images Are More than Pictures, They Are Data. rsna.org https://doi.org/10.1148/radiol.2015151169 (2015) .
9. Hernandez-Boussard, T., Monda, K. L., Crespo, B. C. & Riskin, D. Real world evidence in cardiovascular medicine: ensuring data validity in electronic health record-based studies. OUP Academic https://dx.doi.org/10.1093/jamia/ocz119 (2019) doi:10.1093/jamia/ocz119 .