Artificial Intelligence Model Assists Knee Osteoarthritis Diagnosis via Determination of K-L Grade

<b>Background:</b> Knee osteoarthritis (KOA) affects 37% of individuals aged ≥ 60 years in the national health survey, causing pain, discomfort, and reduced functional independence. <b>Methods:</b> This study aims to automate the assessment of KOA severity by training deep le...

Full description

Saved in:
Bibliographic Details
Main Authors: Joo Chan Choi, Min Young Jeong, Young Jae Kim, Kwang Gi Kim
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/10/1220
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850258092625756160
author Joo Chan Choi
Min Young Jeong
Young Jae Kim
Kwang Gi Kim
author_facet Joo Chan Choi
Min Young Jeong
Young Jae Kim
Kwang Gi Kim
author_sort Joo Chan Choi
collection DOAJ
description <b>Background:</b> Knee osteoarthritis (KOA) affects 37% of individuals aged ≥ 60 years in the national health survey, causing pain, discomfort, and reduced functional independence. <b>Methods:</b> This study aims to automate the assessment of KOA severity by training deep learning models using the Kellgren–Lawrence grading system (class 0~4). A total of 15,000 images were used, with 3000 images collected for each grade. The learning models utilized were DenseNet201, ResNet101, and EfficientNetV2, and their performance in lesion classification was evaluated and compared. Statistical metrics, including accuracy, precision, recall, and F1-score, were employed to assess the feasibility of applying deep learning models for KOA classification. <b>Results:</b> Among these four metrics, DenseNet201 achieved the highest performance, while the ResNet101 model recorded the lowest. DenseNet201 demonstrated the best performance with an overall accuracy of 73%. The model’s accuracy by K-L grade was 80.7% for K-L Grade 0, 53.7% for K-L Grade 1, 72.7% for K-L Grade 2, 75.3% for K-L Grade 3, and 82.7% for K-L Grade 4. The model achieved a precision of 73.2%, a recall of 73%, and an F1-score of 72.7%. <b>Conclusions:</b> These results highlight the potential of deep learning models for assisting specialists in diagnosing the severity of KOA by automatically assigning K-L grades to patient data.
format Article
id doaj-art-230366fb6db142efb3f620e5a4d6afbb
institution OA Journals
issn 2075-4418
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-230366fb6db142efb3f620e5a4d6afbb2025-08-20T01:56:16ZengMDPI AGDiagnostics2075-44182025-05-011510122010.3390/diagnostics15101220Artificial Intelligence Model Assists Knee Osteoarthritis Diagnosis via Determination of K-L GradeJoo Chan Choi0Min Young Jeong1Young Jae Kim2Kwang Gi Kim3Department of Biomedical Engineering, College of Health & Science, Gachon University, Seongnam-si 461701, Republic of KoreaDepartment of Electrical and Computer Engineering, College of Information and Communication Engineering, Sungkyunkwan University, Suwon 16419, Republic of KoreaGachon Biomedical & Convergence Institute, Gachon University Gil Medical Center, Incheon 21565, Republic of KoreaDepartment of Biomedical Engineering, College of Health & Science, Gachon University, Seongnam-si 461701, Republic of Korea<b>Background:</b> Knee osteoarthritis (KOA) affects 37% of individuals aged ≥ 60 years in the national health survey, causing pain, discomfort, and reduced functional independence. <b>Methods:</b> This study aims to automate the assessment of KOA severity by training deep learning models using the Kellgren–Lawrence grading system (class 0~4). A total of 15,000 images were used, with 3000 images collected for each grade. The learning models utilized were DenseNet201, ResNet101, and EfficientNetV2, and their performance in lesion classification was evaluated and compared. Statistical metrics, including accuracy, precision, recall, and F1-score, were employed to assess the feasibility of applying deep learning models for KOA classification. <b>Results:</b> Among these four metrics, DenseNet201 achieved the highest performance, while the ResNet101 model recorded the lowest. DenseNet201 demonstrated the best performance with an overall accuracy of 73%. The model’s accuracy by K-L grade was 80.7% for K-L Grade 0, 53.7% for K-L Grade 1, 72.7% for K-L Grade 2, 75.3% for K-L Grade 3, and 82.7% for K-L Grade 4. The model achieved a precision of 73.2%, a recall of 73%, and an F1-score of 72.7%. <b>Conclusions:</b> These results highlight the potential of deep learning models for assisting specialists in diagnosing the severity of KOA by automatically assigning K-L grades to patient data.https://www.mdpi.com/2075-4418/15/10/1220knee osteoarthritisKellgren–Lawrence grading systemdeep learningDenseNet201ResNet101EfficientNetV2
spellingShingle Joo Chan Choi
Min Young Jeong
Young Jae Kim
Kwang Gi Kim
Artificial Intelligence Model Assists Knee Osteoarthritis Diagnosis via Determination of K-L Grade
Diagnostics
knee osteoarthritis
Kellgren–Lawrence grading system
deep learning
DenseNet201
ResNet101
EfficientNetV2
title Artificial Intelligence Model Assists Knee Osteoarthritis Diagnosis via Determination of K-L Grade
title_full Artificial Intelligence Model Assists Knee Osteoarthritis Diagnosis via Determination of K-L Grade
title_fullStr Artificial Intelligence Model Assists Knee Osteoarthritis Diagnosis via Determination of K-L Grade
title_full_unstemmed Artificial Intelligence Model Assists Knee Osteoarthritis Diagnosis via Determination of K-L Grade
title_short Artificial Intelligence Model Assists Knee Osteoarthritis Diagnosis via Determination of K-L Grade
title_sort artificial intelligence model assists knee osteoarthritis diagnosis via determination of k l grade
topic knee osteoarthritis
Kellgren–Lawrence grading system
deep learning
DenseNet201
ResNet101
EfficientNetV2
url https://www.mdpi.com/2075-4418/15/10/1220
work_keys_str_mv AT joochanchoi artificialintelligencemodelassistskneeosteoarthritisdiagnosisviadeterminationofklgrade
AT minyoungjeong artificialintelligencemodelassistskneeosteoarthritisdiagnosisviadeterminationofklgrade
AT youngjaekim artificialintelligencemodelassistskneeosteoarthritisdiagnosisviadeterminationofklgrade
AT kwanggikim artificialintelligencemodelassistskneeosteoarthritisdiagnosisviadeterminationofklgrade