4 key benefits of applying AI to medical records

Simone Edelmann, PhD

Editor at HealthcareTransformers.com

4 key benefits of applying AI to medical records

21 July 2021 | 5min

Quick Takes

  • Digitalization and the advancement of artificial intelligence are revolutionizing medical health records

  • Key functions and opportunities of AI in electronic medical records include improving productivity and supporting clinical decisions to drive more efficient and personalized care

  • As the healthcare industry embraces technology, the evolution of the data scientist role in healthcare organizations will continue to grow

Patient medical records have long been viewed as inflexible, difficult to use, archaic and costly to configure.1 However, these records are critical to improving the patient experience throughout their care journey. They provide healthcare professionals with the information needed to help make good clinical decisions and evaluate care management plans across the entire continuum of care. Other benefits include ensuring accuracy and timeliness of payment for services performed, and helping to mitigate malpractice risks.2 

Artificial Intelligence (AI) has been a key driver in digital healthcare transformation with digital applications being used to help patients become better decision makers for their own health while driving efficiencies and reducing costs across the healthcare industry.3 But what about AI’s role in medical health records?

Key opportunities of AI in medical records

Improve productivity

AI tools have recently been developed that can help healthcare providers (HCPs) extract clinically-relevant insights from free text housed in medical records or insurance claims, for example.1,4 One such tool, Healthcare Natural Language API, released by Google Cloud, generates a structured data representation of the medical knowledge stored in these data sources for downstream analysis and automation. This extracted information can include:4

  • Medical concepts, such as medications, procedures, and medical conditions
  • Functional features, such as temporal relationships, subjects, and certainty assessments
  • Relations, such as side effects and medication dosage

Despite the headway being made in this field, the challenge for AI-based tools is extracting data in a standardized format that takes into consideration the entire patient journey from a hollistic view. In order to maximize AI in medical records, healthcare organizations are starting to work closely with data scientists to understand what data is relevant and how to generate value from it, which ultimately leads to value for the patient.

Accelerate digital health 

While many physicians express frustration recording patient medical information electronically – indicating that the time it takes to complete entering data is time lost with their patient – they also think that it is the way of the future.

Some hospitals have implemented scribes to sit in on appointments to document the visit while the physician focuses on the patient. AI is coming into play as several companies are working on developing digital scribes – machine-learning algorithms that can take a conversation between a doctor and patient, deconstruct the text and use it to fill in the relevant information in the patient’s electronic medical record (EMR).5

This can also help mitigate the risk of physician burnout and standardize data input, which is a challenge with EMRs. For example, a strawberry allergy may be input into the ‘notes’ field as opposed to ‘allergies’ field, which could result in skewed results.

Improve personalized care

The use of AI in medical records can help identify patterns and perform outcome predictions. Subsequently, this information can be used to tailor specific treatments to an individual, even down to the level of what physician may be best suited to cater to their needs and outcomes that matter most to them. As a practical example, especially during the early phases of the pandemic, patients with pre-existing but non COVID-19 related conditions could be paired with available caregivers based on their data and the outcomes observed across providers. This could potentially help them avoid long waiting times or keep up with their routine health checks in the case that their regular doctor is not available due to office closures. This not only offers improved patient outcomes, but also improves accessibility to care on an individual basis. 

The use of AI can also enable doctors to be alerted to preventative screenings, vaccinations or checkups which takes personalized healthcare to a new level. 

Decision support 

AI-based clinical decision support (CDS) tools are being used to improve care delivery. These tools can analyze large volumes of data to provide diagnostic assistance, treatment guidance and evaluate disease prognosis and progression.7,8,9

Despite the many advantages to CDS tools, due to the vast number of them, their design needs to be carefully implemented to ensure that they do what they are intended to do – Create less work for HCPs and not more. One thing is certain: people from all sectors of the healthcare industry need to be part of the CDS design and implementation process for this to be successful.

AI and the future of medical records

Innovation will continue to advance AI’s role in medical records. It is already being used to analyze large amounts of data to improve productivity, accelerate digital health, improve personalized care and support the clinical decision making process. 

As the healthcare industry embraces technology, the evolution of the data scientist role and the focus on data within healthcare organizations will grow.  Patient experience and outcomes will progressively improve, and this will be partially attributable to the data collected within this valuable resource.

Simone Edelmann, PhD is an editor and contributor at HealthcareTransformers.com. After completing her PhD from the Institute of Biotechnology at the University of Lausanne, Switzerland, she found her passion in medical and scientific communications. She is dedicated to delivering high-quality content on the topic of the future of healthcare to our readers.

References

1 Davenport et al. (2018). Article available from https://hbr.org/2018/12/using-ai-to-improve-electronic-health-records [Accessed July 2021]

2 Thompson. (2016). Article available from https://www.scp-health.com/providers/blog/think-with-your-ink-4-reasons-why-proper-medical-record-documentation-is-vital#:~:text=Clear%20and%20concise%20medical%20record,maintain%20the%20continuum%20of%20care[Accessed November 2020]

3 Licholai. (2020). Article available from https://www.forbes.com/sites/greglicholai/2020/01/14/digital-healthcare-growth-drivers-in-2020/?sh=7f45805b511d [Accessed November 2020]

4. Cloud Healthcare API. Guide documentation available from https://cloud.google.com/healthcare/docs/concepts/nlp#overview [Accessed July 2021]

5. Willyard. (2019). Article available from https://www.nature.com/articles/d41586-019-03848-y [Accessed November 2020]

6. Tang. (2020). Article available from https://ai-med.io/ai-in-medicine/combining-ai-and-electronic-health-records-as-a-new-way-to-pair-patients-and-caregivers/ [Accessed November 2020]

7. Yuan et al. (2020). Int J Med Sci 17, 970–984

8. Mazo et al. (2020). Cancers (Basel) 12, 369

9. Rajwa et al. (2017). IEEE Trans Biomed Eng 64, 1089–1098