Comparison of a Deep Learning Algorithm vs. Humans for Vertebral Heart Scale Measurements in Cats and Dogs Shows a High Degree of Agreement Among Readers.

Emilie Boissady,Jonathan Abbott,Xiajuan Zhu,Hespel Adrien-Maxence,Alois De La Comble

doi:10.3389/fvets.2021.764570

Emilie Boissady, Jonathan Abbott + Show 3 more

Open Access

https://doi.org/10.3389/fvets.2021.764570

Copy DOI

Abstract

Heart disease is a leading cause of death among cats and dogs. Vertebral heart scale (VHS) is one tool to quantify radiographic cardiac enlargement and to predict the occurrence of congestive heart failure. The aim of this study was to evaluate the performance of artificial intelligence (AI) performing VHS measurements when compared with two board-certified specialists. Ground truth consisted of the average of constituent VHS measurements performed by board-certified specialists. Thirty canine and 30 feline thoracic lateral radiographs were evaluated by each operator, using two different methods for determination of the cardiac short axis on dogs' radiographs: the original approach published by Buchanan and the modified approach proposed by the EPIC trial authors, and only Buchanan's method for cats' radiographs. Overall, the VHS calculated by the AI, radiologist, and cardiologist had a high degree of agreement in both canine and feline patients (intraclass correlation coefficient (ICC) = 0.998). In canine patients, when comparing methods used to calculate VHS by specialists, there was also a high degree of agreement (ICC = 0.999). When evaluating specifically the results of the AI VHS vs. the two specialists' readings, the agreement was excellent for both canine (ICC = 0.998) and feline radiographs (ICC = 0.998). Performance of AI trained to locate VHS reference points agreed with manual calculation by specialists in both cats and dogs. Such a computer-aided technique might be an important asset for veterinarians in general practice to limit interobserver variability and obtain more comparable VHS reading over time.

Highlights

Heart disease is a leading cause of death among aging cats and dogs [1]
The widths, lengths, and Vertebral heart scale (VHS) calculated by the artificial intelligence (AI), radiologist, and cardiologist had a high degree of agreement in both canine and feline patients (Table 1)
This study shows that the convolutional neural networks (CNNs) developed for this project provided a VHS measurement in both and cats, which has a statistically high agreement with the measurements obtained by a cardiologist and a radiologist

Summary

Introduction

Heart disease is a leading cause of death among aging cats and dogs [1]. While echocardiography remains the standard imaging modality for evaluation of cardiac diseases [2], radiology plays an important part in their diagnosis and management. As heart disease is frequently associated with enlargement of the cardiac silhouette, this low-cost and widely available exam can be used first as a screening method when a patient is presented with equivocal clinical signs (cough, fatigability, etc.) [3, 4]. The width is considered enlarged if >2.5–3.5 intercostal spaces [5, 6] This semiquantitative method lacks precision, especially in cases of mild-to-moderate cardiomegaly, making the diagnosis arduous for non-specialist veterinarians [7]. This is further accentuated by many factors that may interfere with cardiac size evaluation, such as thoracic conformation (for example, barrel-chested dogs physiologically exceed the ranges presented above) or respiratory cycle [6]

Objectives

Methods

Results