An anemia screening tool based on deep learning with conjunctiva images
Objective To explore the application of deep learning in automatic classification of anemia with conjunctival images as input. Methods The conjunctival images of 284 patients undergoing elective surgery in the Department of Anesthesiology of the First Affiliated Hospital of Army Medical University f...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Journal of Army Medical University
2023-04-01
|
| Series: | 陆军军医大学学报 |
| Subjects: | |
| Online Access: | http://aammt.tmmu.edu.cn/html/202301049.htm |
| _version_ | 1797832779515297792 |
|---|---|
| author | HU Xiaoyan LI Haoyang LIU Xiang LI Yujie TAN Lifang |
| author_facet | HU Xiaoyan LI Haoyang LIU Xiang LI Yujie TAN Lifang |
| author_sort | HU Xiaoyan |
| collection | DOAJ |
| description | Objective To explore the application of deep learning in automatic classification of anemia with conjunctival images as input. Methods The conjunctival images of 284 patients undergoing elective surgery in the Department of Anesthesiology of the First Affiliated Hospital of Army Medical University from March 18 to April 26, 2021 were collected and analyzed prospectively. The images divided into 2 types: normal and anemia according to the corresponding hemoglobin concentration. Four deep learning algorithms, including InceptionV3, ResNet50V2, EfficientNetV2B0 and DenseNet121, were used to construct a prediction model for anemia. The performance of the model was evaluated by receiver operating characteristic (ROC) curve with accuracy, sensitivity, specificity, positive predictive value and negative predictive value. Results The area under ROC curve (AUC) was 0.709 (95%CI: 0.643~0.769), 0.661 (95%CI: 0.594~0.725), 0.670 (95%CI: 0.603~0.733), and 0.695 (95%CI: 0.628~0.756), respectively for the 4 deep learning algorithms. The InceptionV3 model showed superior predictive performance on the test set, with an AUC value of 0.709 (95%CI: 0.643~0.769), an accuracy of 0.695, a sensitivity of 0.750, a specificity of 0.412, a positive predictive value of 0.707 and a negative predictive value of 0.629. Based on the optimal algorithm, a network service application which can be used for online prediction of anemia was developed (http://150.158.58.4). Conclusion Our model, which is established based on deep learning algorithm with conjunctiva image as input, has a good performance on fast and automatic prediction for anemia. The InceptionV3model has better comprehensive prediction performance.
|
| first_indexed | 2024-04-09T14:14:18Z |
| format | Article |
| id | doaj.art-8f2521c4cf184b50829f8d6a68280542 |
| institution | Directory Open Access Journal |
| issn | 2097-0927 |
| language | zho |
| last_indexed | 2024-04-09T14:14:18Z |
| publishDate | 2023-04-01 |
| publisher | Editorial Office of Journal of Army Medical University |
| record_format | Article |
| series | 陆军军医大学学报 |
| spelling | doaj.art-8f2521c4cf184b50829f8d6a682805422023-05-05T23:54:46ZzhoEditorial Office of Journal of Army Medical University陆军军医大学学报2097-09272023-04-0145874675210.16016/j.2097-0927.202301049An anemia screening tool based on deep learning with conjunctiva imagesHU Xiaoyan0LI Haoyang1LIU Xiang2LI Yujie3TAN Lifang4Department of Anesthesiology, First Affiliated Hospital, Army Medical University, Army Medical University (Third Military Medical University), Chongqing, 400038Regiment Five, Basical Medicine College, Army Medical University, Army Medical University (Third Military Medical University), Chongqing, 400038Department of Anesthesiology, First Affiliated Hospital, Army Medical University, Army Medical University (Third Military Medical University), Chongqing, 400038Department of Anesthesiology, First Affiliated Hospital, Army Medical University, Army Medical University (Third Military Medical University), Chongqing, 400038Department of Anesthesiology, First Affiliated Hospital, Army Medical University, Army Medical University (Third Military Medical University), Chongqing, 400038Objective To explore the application of deep learning in automatic classification of anemia with conjunctival images as input. Methods The conjunctival images of 284 patients undergoing elective surgery in the Department of Anesthesiology of the First Affiliated Hospital of Army Medical University from March 18 to April 26, 2021 were collected and analyzed prospectively. The images divided into 2 types: normal and anemia according to the corresponding hemoglobin concentration. Four deep learning algorithms, including InceptionV3, ResNet50V2, EfficientNetV2B0 and DenseNet121, were used to construct a prediction model for anemia. The performance of the model was evaluated by receiver operating characteristic (ROC) curve with accuracy, sensitivity, specificity, positive predictive value and negative predictive value. Results The area under ROC curve (AUC) was 0.709 (95%CI: 0.643~0.769), 0.661 (95%CI: 0.594~0.725), 0.670 (95%CI: 0.603~0.733), and 0.695 (95%CI: 0.628~0.756), respectively for the 4 deep learning algorithms. The InceptionV3 model showed superior predictive performance on the test set, with an AUC value of 0.709 (95%CI: 0.643~0.769), an accuracy of 0.695, a sensitivity of 0.750, a specificity of 0.412, a positive predictive value of 0.707 and a negative predictive value of 0.629. Based on the optimal algorithm, a network service application which can be used for online prediction of anemia was developed (http://150.158.58.4). Conclusion Our model, which is established based on deep learning algorithm with conjunctiva image as input, has a good performance on fast and automatic prediction for anemia. The InceptionV3model has better comprehensive prediction performance. http://aammt.tmmu.edu.cn/html/202301049.htmdeep learninganemiaconjunctivanon-invasiveness |
| spellingShingle | HU Xiaoyan LI Haoyang LIU Xiang LI Yujie TAN Lifang An anemia screening tool based on deep learning with conjunctiva images 陆军军医大学学报 deep learning anemia conjunctiva non-invasiveness |
| title | An anemia screening tool based on deep learning with conjunctiva images |
| title_full | An anemia screening tool based on deep learning with conjunctiva images |
| title_fullStr | An anemia screening tool based on deep learning with conjunctiva images |
| title_full_unstemmed | An anemia screening tool based on deep learning with conjunctiva images |
| title_short | An anemia screening tool based on deep learning with conjunctiva images |
| title_sort | anemia screening tool based on deep learning with conjunctiva images |
| topic | deep learning anemia conjunctiva non-invasiveness |
| url | http://aammt.tmmu.edu.cn/html/202301049.htm |
| work_keys_str_mv | AT huxiaoyan ananemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT lihaoyang ananemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT liuxiang ananemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT liyujie ananemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT tanlifang ananemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT huxiaoyan anemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT lihaoyang anemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT liuxiang anemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT liyujie anemiascreeningtoolbasedondeeplearningwithconjunctivaimages AT tanlifang anemiascreeningtoolbasedondeeplearningwithconjunctivaimages |