# Patient Self-Scanning for Lung Ultrasound: A Prospective Observational Study on the Feasibility and Diagnostic Accuracy of a Telemedicine Protocol

## TL;DR

This study shows that patients can successfully perform their own lung ultrasounds with brief video instructions, supporting the use of telemedicine for remote care.

## Contribution

Demonstrates the feasibility and diagnostic potential of patient-performed lung ultrasound for telemedicine applications.

## Key findings

- Patients achieved high comfort and willingness to perform lung ultrasound with minimal instruction.
- Most patient-obtained images were interpretable, though provider images scored higher overall.
- Inter-rater reliability for image quality was good, indicating consistent evaluation by trained physicians.

## Abstract

Objective: To evaluate the feasibility and diagnostic accuracy of patient-performed lung ultrasound (P-PLUS) for telemedicine purposes.

Methods: This prospective observation study included patients over 18 years old who presented to a tertiary care hospital's emergency department. Patients were provided a five-minute instructional video on a US protocol of four lung zones and then performed the protocol while being monitored by a study investigator. The physician sonographer subsequently repeated the protocol. Two emergency physicians with US fellowship training blindly reviewed and independently rated image quality on a scale of one to five, with a score of three or more considered interpretable. Inter-rater reliability was estimated using the intraclass correlation coefficient. Wilcoxon-Mann-Whitney tests and chi-square tests were used to compare group differences.

Results: A total of 56 patients (45% female) were enrolled, and 417 clips were analyzed. Ten (18%) participants worked in the medical field, and 44 (79%) had at least some college education. Forty (71%) regularly used technology at work, 52 (93%) had internet access at home, and the same number had access to smartphones. Patients reported high comfort in performing self-LUS (median score: 4, interquartile range (IQR): 3.5-5) and high willingness to perform US acquisition again in the future (median score: 4, IQR: 4-5). The proportion of interpretable images was similar between the two groups except for the left hemidiaphragm (90% of provider-obtained images were interpretable vs. 45% of patient-obtained images, P = 0.002). The majority of patient-obtained images were scored between three and four and classified as interpretable. Mean image scores were significantly higher for provider-obtained images (P < 0.05). Inter-rater reliability between the two raters was good (intraclass correlation coefficient (ICC)= 0.80, 95% CI 0.76-0.84).

Conclusion: Patients can independently obtain interpretable LUS images in all views with minimal video-based instruction. The ability of patients to obtain interpretable LUS images with minimal tele-guidance, as well as their high levels of comfort and willingness to perform the procedure, support the potential use of P-PLUS in home-based and remote patient care.