FP-CNN : Fuzzy pooling-based convolutional neural network for lung ultrasound image classification with explainable AI
Copyright © 2023 The Author(s). Published by Elsevier Ltd.. All rights reserved..
The COVID-19 pandemic wreaks havoc on healthcare systems all across the world. In pandemic scenarios like COVID-19, the applicability of diagnostic modalities is crucial in medical diagnosis, where non-invasive ultrasound imaging has the potential to be a useful biomarker. This research develops a computer-assisted intelligent methodology for ultrasound lung image classification by utilizing a fuzzy pooling-based convolutional neural network FP-CNN with underlying evidence of particular decisions. The fuzzy-pooling method finds better representative features for ultrasound image classification. The FPCNN model categorizes ultrasound images into one of three classes: covid, disease-free (normal), and pneumonia. Explanations of diagnostic decisions are crucial to ensure the fairness of an intelligent system. This research has used Shapley Additive Explanation (SHAP) to explain the prediction of the FP-CNN models. The prediction of the black-box model is illustrated using the SHAP explanation of the intermediate layers of the black-box model. To determine the most effective model, we have tested different state-of-the-art convolutional neural network architectures with various training strategies, including fine-tuned models, single-layer fuzzy pooling models, and fuzzy pooling at all pooling layers. Among different architectures, the Xception model with all pooling layers having fuzzy pooling achieves the best classification results of 97.2% accuracy. We hope our proposed method will be helpful for the clinical diagnosis of covid-19 from lung ultrasound (LUS) images.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:165 |
|---|---|
| Enthalten in: |
Computers in biology and medicine - 165(2023) vom: 01. Okt., Seite 107407 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Hasan, Md Mahmodul [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
COVID-19 diagnosis |
|---|
| Anmerkungen: |
Date Completed 27.09.2023 Date Revised 27.09.2023 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1016/j.compbiomed.2023.107407 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM361757654 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM361757654 | ||
| 003 | DE-627 | ||
| 005 | 20231226085820.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.compbiomed.2023.107407 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1205.xml |
| 035 | |a (DE-627)NLM361757654 | ||
| 035 | |a (NLM)37678140 | ||
| 035 | |a (PII)S0010-4825(23)00872-7 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Hasan, Md Mahmodul |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a FP-CNN |b Fuzzy pooling-based convolutional neural network for lung ultrasound image classification with explainable AI |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 27.09.2023 | ||
| 500 | |a Date Revised 27.09.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Copyright © 2023 The Author(s). Published by Elsevier Ltd.. All rights reserved. | ||
| 520 | |a The COVID-19 pandemic wreaks havoc on healthcare systems all across the world. In pandemic scenarios like COVID-19, the applicability of diagnostic modalities is crucial in medical diagnosis, where non-invasive ultrasound imaging has the potential to be a useful biomarker. This research develops a computer-assisted intelligent methodology for ultrasound lung image classification by utilizing a fuzzy pooling-based convolutional neural network FP-CNN with underlying evidence of particular decisions. The fuzzy-pooling method finds better representative features for ultrasound image classification. The FPCNN model categorizes ultrasound images into one of three classes: covid, disease-free (normal), and pneumonia. Explanations of diagnostic decisions are crucial to ensure the fairness of an intelligent system. This research has used Shapley Additive Explanation (SHAP) to explain the prediction of the FP-CNN models. The prediction of the black-box model is illustrated using the SHAP explanation of the intermediate layers of the black-box model. To determine the most effective model, we have tested different state-of-the-art convolutional neural network architectures with various training strategies, including fine-tuned models, single-layer fuzzy pooling models, and fuzzy pooling at all pooling layers. Among different architectures, the Xception model with all pooling layers having fuzzy pooling achieves the best classification results of 97.2% accuracy. We hope our proposed method will be helpful for the clinical diagnosis of covid-19 from lung ultrasound (LUS) images | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, Non-U.S. Gov't | |
| 650 | 4 | |a COVID-19 diagnosis | |
| 650 | 4 | |a Explainable artificial intelligence (XAI) | |
| 650 | 4 | |a Fuzzy pooling | |
| 650 | 4 | |a Ultrasound image classification | |
| 700 | 1 | |a Hossain, Muhammad Minoar |e verfasserin |4 aut | |
| 700 | 1 | |a Rahman, Mohammad Motiur |e verfasserin |4 aut | |
| 700 | 1 | |a Azad, Akm |e verfasserin |4 aut | |
| 700 | 1 | |a Alyami, Salem A |e verfasserin |4 aut | |
| 700 | 1 | |a Moni, Mohammad Ali |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Computers in biology and medicine |d 1970 |g 165(2023) vom: 01. Okt., Seite 107407 |w (DE-627)NLM000382272 |x 1879-0534 |7 nnns |
| 773 | 1 | 8 | |g volume:165 |g year:2023 |g day:01 |g month:10 |g pages:107407 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.compbiomed.2023.107407 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 165 |j 2023 |b 01 |c 10 |h 107407 | ||