Ambient virtual scribes: Mutuo Health’s AutoScribe as a case study of artificial intelligence-based technology

Abstract

Studies show that clinicians are increasingly burning out in large part from the clerical burden associated with using Electronic Medical Record (EMR) systems. At the same time, recently developed health data analytic algorithms struggle with poor quality free-text entered data in these systems. We developed AutoScribe using artificial intelligence–based natural language processing tools to automate these clerical tasks and to output high-quality EMR data. In this article, we describe the benefits and drawbacks of our technology. Furthermore, we describe how we are positioning our company’s culture within the existing healthcare system and suggest steps leaders of the system should consider in order to ensure that potentially transformative artificial intelligence–based technologies like ours are optimally adopted.

Get full access to this article

View all access options for this article.

References

Institute of Medicine. Digital Infrastructure for the Learning Health System: The Foundation for Continuous Improvement in Health and Health Care. Washington, DC: Institute of Medicine; 2011. Available at: https://www.ncbi.nlm.nih.gov/books/NBK83569/pdf/Bookshelf_NBK83569.pdf.

Sinsky

Colligan

, et al. Allocation of physician time in ambulatory practice: a time and motion study in 4 specialties. Ann Intern Med. 2016;165(11):753–760. doi:10.7326/M16-0961.

Gawande

. Why doctors hate their computers. 2019. Available at: https://www.newyorker.com/magazine/2018/11/12/why-doctors-hate-their-computers.

Can We Talk About Physician Mental Health?

UofTMed Magazine. 2017. Available at: https://uoftmedmagazine.utoronto.ca/2017/summer/.

Rheaume

. Burnout and mental illness among Canadian physicians. UBC Med J. 2016;8(1):48–49. Available at: http://med-fom-ubcmj.sites.olt.ubc.ca/files/2017/03/v8i1-news3.pdf.

Bodenheimer

Sinsky

. From triple to quadruple aim: care of the patient requires care of the provider. Ann Fam Med. 2014;12(6):573–576. doi:10.1370/afm.1713.

Weiskopf

Weng

. Methods and dimensions of electronic health record data quality assessment: enabling reuse for clinical research. J Am Med Inform Assoc. 2013;20(1):144–151. Available at: http://jamia.oxfordjournals.org/content/20/1/144.abstract.

Topol

. Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again. New York, NY: Basic Books; 2019.

LeCun

Bengio

Hinton

. Deep learning. Nature. 2015;521(7553):436–444. doi:10.1038/nature14539.

10.

Mikolov

Sutskever

Chen

Corrado

Dean

Distributed Representations of Words and Phrases and Their Compositionality. In: Proceedings of the 26^th International Conference on Neural Information Processing Systems Volume 2 (NIPS’13) 2013; 2013:3111–3119.

11.

Singh

. Natural language processing for information extraction. Computation and Language; arXiv.org; 2018. Available at: https://arxiv.org/abs/1807.02383.

12.

Faiza Khan

Jeblee

Crampton

Mamdani

Rudzicz

. AutoScribe: extracting clinically pertinent information from patient-clinician dialogues. Med Info. 2019;264:1512–1513. doi:10.3233/SHTI190510.

13.

Heaton

Nestler

Barry

, et al. A time-driven activity-based costing analysis of emergency department scribes. Mayo Clin Proc Innov Qual Outcomes. 2019;3(1):30–34.

14.

What is the Patient-Centered Medical Home? (PCMH) | ACP. 2019. Available at: https://www.acponline.org/practice-resources/business-resources/payment/delivery-and-payment-models/patient-centered-medical-home/understanding-the-patient-centered-medical-home/what-is-the-patient-centered-medical-home.

15.

Holzinger

Biemann

Pattichis

Kell

. What do we need to build explainable AI systems for the medical domain? arXiv:1712.09923v1 [cs.AI]; 2017.

16.

The FDA wants to regulate machine learning in healthcare. 2019. Available at: https://www.technologyreview.com/f/613264/the-fda-wants-to-regulate-machine-learning-in-healthcare/.