Pulmonary Hypertension Classification using Artificial Intelligence and Chest X-Ray:ATA AI STUDY-1
Other publications > Other · 2023
Authors: Tarık Kıvrak, Burcu YAĞMUR, Hilal Erken, Derya Kocakaya, Turker Tuncer, Şengül Doğan, Orhan Yaman, Umit Yasar Sinan, Sena Sert Sekerci, Cagri Yayla, Ufuk Iyigun, Mehmet Kis, Ozkan Karaca, Emrah Yesil, Elif Ilkay Yuce Ersoy, Bahar Tekin Tak, Ahmet Oz, Mehmet Kaplan, Zeynep Ulutas, Gamze Yeter Aslan, Nihan Kahya Eren, Fatma Nihan Turhan Caglar, Hatice Solmaz, Ozge Ozden, Hakan Gunes, Umut Kocabas, Mustafa Yenercag, Omer Isık, Cem Yesilkaya, Ali Nail Kaya, Sefa Erdi Omur, Anil Sahin, Erdal In, Nurcan Kırıcı Berber, Cigdem Ileri Dogan, Fatih Poyraz, Emin Erdem Kaya, Ayca Gumusdag, Omer Kumet, Hakki Kaya, Remzi Sarikaya, Seda Turkan Tan, Hidayet Ozan Arabaci, Rengin Cetin Guvenc, Mehtap Yeni, Burcak Kılıckıran Avci, Dilek Cicek Yilmaz, Ahmet Celik, Berkay Ekici, Aycan Fahri Erkan, Veysel Ozgur Baris, Taner Seker, Ferit Böyük, Mehmet Mustafa Can, Hasan Gungor, Hakki Simsek, Bedrettin Yildizeli, Mehmet Ali Kobat, Mehmet Akbulut, Mehdi ZOGHİ, Omer Kozan
Other, International, 2023,
Summary
Abstract An accurate diagnosis of pulmonary hypertension (PH) is crucial to ensure that patients receive timely treatment. One of the used imaging models to detect pulmonary hypertension is the X-ray. Therefore, a new automated PH-type classification model has been presented to depict the separation ability of deep learning for PH types. We retrospectively enrolled 6642 images of patients with PH and the control group. A new X-ray image dataset was collected from a multicentre in this work. A transfer learning-based image classification model has been presented in classifying PH types. Our proposed model was applied to the collected dataset, and this dataset contains six categories (five PH and a non-PH). The presented deep feature engineering (computer vision) model attained 86.14% accuracy on this dataset. According to the extracted ROC curve, the average area under the curve rate has been calculated at 0.945. Therefore, we believe that our proposed model can easily separate PH and non-PH X-ray images.