LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications
This study presents a comprehensive analysis of utilizing TensorFlow Lite on mobile phones for the on-edge medical diagnosis of lung diseases. This paper focuses on the technical deployment of various deep learning architectures to classify nine respiratory system diseases using X-ray imagery. We pr...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | , |
| التنسيق: | article |
| منشور في: |
2024
|
| الموضوعات: | |
| الوصول للمادة أونلاين: | https://hdl.handle.net/11073/25544 |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| _version_ | 1864513432400691200 |
|---|---|
| author | Aldamani, Raghad |
| author2 | Abuhani, Diaa Addeen Shanableh, Tamer |
| author2_role | author author |
| author_facet | Aldamani, Raghad Abuhani, Diaa Addeen Shanableh, Tamer |
| author_role | author |
| dc.creator.none.fl_str_mv | Aldamani, Raghad Abuhani, Diaa Addeen Shanableh, Tamer |
| dc.date.none.fl_str_mv | 2024-07-01T05:34:10Z 2024-07-01T05:34:10Z 2024 |
| dc.format.none.fl_str_mv | application/pdf |
| dc.identifier.none.fl_str_mv | Aldamani, R.; Abuhani, D.A.; Shanableh, T. LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications. Algorithms 2024, 17, 280. https://doi.org/10.3390/a17070280 1999-4893 https://hdl.handle.net/11073/25544 10.3390/a17070280 |
| dc.language.none.fl_str_mv | en_US |
| dc.publisher.none.fl_str_mv | MDPI |
| dc.relation.none.fl_str_mv | https://doi.org/10.3390/a17070280 |
| dc.subject.none.fl_str_mv | Medical diagnosis CNN image classification Model quantization On-edge image classification |
| dc.title.none.fl_str_mv | LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications |
| dc.type.none.fl_str_mv | Peer-Reviewed Published version info:eu-repo/semantics/publishedVersion info:eu-repo/semantics/article |
| description | This study presents a comprehensive analysis of utilizing TensorFlow Lite on mobile phones for the on-edge medical diagnosis of lung diseases. This paper focuses on the technical deployment of various deep learning architectures to classify nine respiratory system diseases using X-ray imagery. We propose a simple deep learning architecture that experiments with six different convolutional neural networks. Various quantization techniques are employed to convert the classification models into TensorFlow Lite, including post-classification quantization with floating point 16 bit representation, integer quantization with representative data, and quantization-aware training. This results in a total of 18 models suitable for on-edge deployment for the classification of lung diseases. We then examine the generated models in terms of model size reduction, accuracy, and inference time. Our findings indicate that the quantization-aware training approach demonstrates superior optimization results, achieving an average model size reduction of 75.59%. Among many CNNs, MobileNetV2 exhibited the highest performance-to-size ratio, with an average accuracy loss of 4.1% across all models using the quantization-aware training approach. In terms of inference time, TensorFlow Lite with integer quantization emerged as the most efficient technique, with an average improvement of 1.4 s over other conversion approaches. Our best model, which used EfficientNetB2, achieved an F1-Score of approximately 98.58%, surpassing state-of-the-art performance on the X-ray lung diseases dataset in terms of accuracy, specificity, and sensitivity. The model experienced an F1 loss of around 1% using quantization-aware optimization. The study culminated in the development of a consumer-ready app, with TensorFlow Lite models tailored to mobile devices. |
| format | article |
| id | aus_f10d772b9c997f722d339b6b8ca64798 |
| identifier_str_mv | Aldamani, R.; Abuhani, D.A.; Shanableh, T. LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications. Algorithms 2024, 17, 280. https://doi.org/10.3390/a17070280 1999-4893 10.3390/a17070280 |
| language_invalid_str_mv | en_US |
| network_acronym_str | aus |
| network_name_str | aus |
| oai_identifier_str | oai:repository.aus.edu:11073/25544 |
| publishDate | 2024 |
| publisher.none.fl_str_mv | MDPI |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| spelling | LungVision: X-ray Imagery Classification for On-Edge Diagnosis ApplicationsAldamani, RaghadAbuhani, Diaa AddeenShanableh, TamerMedical diagnosisCNN image classificationModel quantizationOn-edge image classificationThis study presents a comprehensive analysis of utilizing TensorFlow Lite on mobile phones for the on-edge medical diagnosis of lung diseases. This paper focuses on the technical deployment of various deep learning architectures to classify nine respiratory system diseases using X-ray imagery. We propose a simple deep learning architecture that experiments with six different convolutional neural networks. Various quantization techniques are employed to convert the classification models into TensorFlow Lite, including post-classification quantization with floating point 16 bit representation, integer quantization with representative data, and quantization-aware training. This results in a total of 18 models suitable for on-edge deployment for the classification of lung diseases. We then examine the generated models in terms of model size reduction, accuracy, and inference time. Our findings indicate that the quantization-aware training approach demonstrates superior optimization results, achieving an average model size reduction of 75.59%. Among many CNNs, MobileNetV2 exhibited the highest performance-to-size ratio, with an average accuracy loss of 4.1% across all models using the quantization-aware training approach. In terms of inference time, TensorFlow Lite with integer quantization emerged as the most efficient technique, with an average improvement of 1.4 s over other conversion approaches. Our best model, which used EfficientNetB2, achieved an F1-Score of approximately 98.58%, surpassing state-of-the-art performance on the X-ray lung diseases dataset in terms of accuracy, specificity, and sensitivity. The model experienced an F1 loss of around 1% using quantization-aware optimization. The study culminated in the development of a consumer-ready app, with TensorFlow Lite models tailored to mobile devices.MDPI2024-07-01T05:34:10Z2024-07-01T05:34:10Z2024Peer-ReviewedPublished versioninfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfAldamani, R.; Abuhani, D.A.; Shanableh, T. LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications. Algorithms 2024, 17, 280. https://doi.org/10.3390/a170702801999-4893https://hdl.handle.net/11073/2554410.3390/a17070280en_UShttps://doi.org/10.3390/a17070280oai:repository.aus.edu:11073/255442024-08-22T12:07:10Z |
| spellingShingle | LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications Aldamani, Raghad Medical diagnosis CNN image classification Model quantization On-edge image classification |
| status_str | publishedVersion |
| title | LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications |
| title_full | LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications |
| title_fullStr | LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications |
| title_full_unstemmed | LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications |
| title_short | LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications |
| title_sort | LungVision: X-ray Imagery Classification for On-Edge Diagnosis Applications |
| topic | Medical diagnosis CNN image classification Model quantization On-edge image classification |
| url | https://hdl.handle.net/11073/25544 |