Quan Zhang1,2, Zhiang Liu3, Jiaxu Li4, Guohua Liu1,2. 1. College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, People's Republic of China. 2. Tianjin Key Laboratory of Optoelectronic Sensor and Sensing Network Technology, Nankai University, Tianjin 300350, People's Republic of China. 3. College of Computer Science, Nankai University, Tianjin 300350, People's Republic of China. 4. The Second Affiliated Hospital of Harbin Medical University, Department of Plastic and Cosmetic Surgery, Harbin, Heilongjiang, 150081, People's Republic of China.
Abstract
PURPOSE: Diabetic Macular Edema has been one of the research hotspots all over the world. But as the global population continues to grow, the number of OCT images requiring manual analysis is becoming increasingly unaffordable. Medical images are often fuzzy due to the inherent physical processes of acquiring them. It is difficult for traditional algorithms to use low-quality data. And traditional algorithms usually only provide diagnostic results, which makes the reliability and interpretability of the model face challenges. To solve problem above, we proposed a more intuitive and robust diagnosis model with self-enhancement ability and clinical triage patients' ability. METHODS: We used 38,057 OCT images (Drusen, DME, CNV and Normal) to establish and evaluate the model. All data are OCT images of fundus retina. There were 37,457 samples in the training dataset and 600 samples in the validation dataset. In order to diagnose these images accurately, we propose a multiscale transfer learning algorithm. Firstly, the sample is sent to the automatic self-enhancement module for edge detection and enhancement. Then, the processed data are sent to the image diagnosis module to determine the disease type. This process makes more data more effective and can be accurately classified. Finally, we calculated the accuracy, precision, sensitivity and specificity of the model, and verified the performance of the model from the perspective of clinical application. RESULTS: The model proposed in this paper can provide the diagnosis results and display the detection targets more intuitively. The model reached 94.5% accuracy, 97.2% precision, 97.7% sensitivity and 97% specificity in the independent testing dataset. CONCLUSION: Comparing the performance of relevant work and ablation test, our model achieved relatively good performance. It is proved that the model proposed in this paper has a stronger ability to recognize diseases even in the face of low-quality images. Experiment results also demonstrate its clinical referral capability. It can reduce the workload of medical staff and save the precious time of patients.
PURPOSE: Diabetic Macular Edema has been one of the research hotspots all over the world. But as the global population continues to grow, the number of OCT images requiring manual analysis is becoming increasingly unaffordable. Medical images are often fuzzy due to the inherent physical processes of acquiring them. It is difficult for traditional algorithms to use low-quality data. And traditional algorithms usually only provide diagnostic results, which makes the reliability and interpretability of the model face challenges. To solve problem above, we proposed a more intuitive and robust diagnosis model with self-enhancement ability and clinical triage patients' ability. METHODS: We used 38,057 OCT images (Drusen, DME, CNV and Normal) to establish and evaluate the model. All data are OCT images of fundus retina. There were 37,457 samples in the training dataset and 600 samples in the validation dataset. In order to diagnose these images accurately, we propose a multiscale transfer learning algorithm. Firstly, the sample is sent to the automatic self-enhancement module for edge detection and enhancement. Then, the processed data are sent to the image diagnosis module to determine the disease type. This process makes more data more effective and can be accurately classified. Finally, we calculated the accuracy, precision, sensitivity and specificity of the model, and verified the performance of the model from the perspective of clinical application. RESULTS: The model proposed in this paper can provide the diagnosis results and display the detection targets more intuitively. The model reached 94.5% accuracy, 97.2% precision, 97.7% sensitivity and 97% specificity in the independent testing dataset. CONCLUSION: Comparing the performance of relevant work and ablation test, our model achieved relatively good performance. It is proved that the model proposed in this paper has a stronger ability to recognize diseases even in the face of low-quality images. Experiment results also demonstrate its clinical referral capability. It can reduce the workload of medical staff and save the precious time of patients.
Authors: Liang Liu; Beth Edmunds; Hana L Takusagawa; Shandiz Tehrani; Lorinna H Lombardi; John C Morrison; Yali Jia; David Huang Journal: Am J Ophthalmol Date: 2019-06-03 Impact factor: 5.258
Authors: Arthur A Bergen; Swati Arya; Céline Koster; Matthew G Pilgrim; Dagmara Wiatrek-Moumoulidis; Peter J van der Spek; Stefanie M Hauck; Camiel J F Boon; Eszter Emri; Alan J Stewart; Imre Lengyel Journal: Prog Retin Eye Res Date: 2018-12-17 Impact factor: 21.198
Authors: N H Cho; J E Shaw; S Karuranga; Y Huang; J D da Rocha Fernandes; A W Ohlrogge; B Malanda Journal: Diabetes Res Clin Pract Date: 2018-02-26 Impact factor: 5.602
Authors: Anne Louise Askou; Sidsel Alsing; Josephine N E Benckendorff; Andreas Holmgaard; Jacob Giehm Mikkelsen; Lars Aagaard; Toke Bek; Thomas J Corydon Journal: Mol Ther Nucleic Acids Date: 2019-02-02
Authors: Yihe Wang; Jason Riordon; Tian Kong; Yi Xu; Brian Nguyen; Junjie Zhong; Jae Bem You; Alexander Lagunov; Thomas G Hannam; Keith Jarvi; David Sinton Journal: Adv Sci (Weinh) Date: 2019-05-24 Impact factor: 16.806
Authors: Junerlyn Agua-Agum; Benedetta Allegranzi; Archchun Ariyarajah; R Bruce Aylward; Isobel M Blake; Philippe Barboza; Daniel Bausch; Richard J Brennan; Peter Clement; Pasqualina Coffey; Anne Cori; Christl A Donnelly; Ilaria Dorigatti; Patrick Drury; Kara Durski; Christopher Dye; Tim Eckmanns; Neil M Ferguson; Christophe Fraser; Erika Garcia; Tini Garske; Alex Gasasira; Céline Gurry; Esther Hamblion; Wes Hinsley; Robert Holden; David Holmes; Stéphane Hugonnet; Giovanna Jaramillo Gutierrez; Thibaut Jombart; Edward Kelley; Ravi Santhana; Nuha Mahmoud; Harriet L Mills; Yasmine Mohamed; Emmanuel Musa; Dhamari Naidoo; Gemma Nedjati-Gilani; Emily Newton; Ian Norton; Pierre Nouvellet; Devin Perkins; Mark Perkins; Steven Riley; Dirk Schumacher; Anita Shah; Minh Tang; Olivia Varsaneux; Maria D Van Kerkhove Journal: N Engl J Med Date: 2016-08-11 Impact factor: 91.245