CLASSIFICATION OF CHEST RADIOGRAPHS USING NOVEL ANOMALOUS SALIENCY MAP AND DEEP CONVOLUTIONAL NEURAL NETWORK
The rapid advancement in pattern recognition via the deep learning method has made it possible to develop an autonomous medical image classification system. This system has proven robust and accurate in classifying most pathological features found in a medical image, such as airspace opacity, mass,...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IIUM Press, International Islamic University Malaysia
2021-07-01
|
| Series: | International Islamic University Malaysia Engineering Journal |
| Subjects: | |
| Online Access: | https://journals.iium.edu.my/ejournal/index.php/iiumej/article/view/1752 |
| _version_ | 1818563570608111616 |
|---|---|
| author | Mohd Adli Md Ali Mohd Radhwan Abidin Nik Arsyad Nik Muhamad Affendi Hafidzul Abdullah Daaniyal R. Rosman Nu'man Barud'din Faiz Kemi Farid Hayati |
| author_facet | Mohd Adli Md Ali Mohd Radhwan Abidin Nik Arsyad Nik Muhamad Affendi Hafidzul Abdullah Daaniyal R. Rosman Nu'man Barud'din Faiz Kemi Farid Hayati |
| author_sort | Mohd Adli Md Ali |
| collection | DOAJ |
| description | The rapid advancement in pattern recognition via the deep learning method has made it possible to develop an autonomous medical image classification system. This system has proven robust and accurate in classifying most pathological features found in a medical image, such as airspace opacity, mass, and broken bone. Conventionally, this system takes routine medical images with minimum pre-processing as the model's input; in this research, we investigate if saliency maps can be an alternative model input. Recent research has shown that saliency maps' application increases deep learning model performance in image classification, object localization, and segmentation. However, conventional bottom-up saliency map algorithms regularly failed to localize salient or pathological anomalies in medical images. This failure is because most medical images are homogenous, lacking color, and contrast variant. Therefore, we also introduce the Xenafas algorithm in this paper. The algorithm creates a new kind of anomalous saliency map called the Intensity Probability Mapping and Weighted Intensity Probability Mapping. We tested the proposed saliency maps on five deep learning models based on common convolutional neural network architecture. The result of this experiment showed that using the proposed saliency map over regular radiograph chest images increases the sensitivity of most models in identifying images with air space opacities. Using the Grad-CAM algorithm, we showed how the proposed saliency map shifted the model attention to the relevant region in chest radiograph images. While in the qualitative study, it was found that the proposed saliency map regularly highlights anomalous features, including foreign objects and cardiomegaly. However, it is inconsistent in highlighting masses and nodules.
ABSTRAK: Perkembangan pesat sistem pengecaman corak menggunakan kaedah pembelajaran mendalam membolehkan penghasilan sistem klasifikasi gambar perubatan secara automatik. Sistem ini berupaya menilai secara tepat jika terdapat tanda-tanda patologi di dalam gambar perubatan seperti kelegapan ruang udara, jisim dan tulang patah. Kebiasaannya, sistem ini akan mengambil gambar perubatan dengan pra-pemprosesan minimum sebagai input. Kajian ini adalah tentang potensi peta salien dapat dijadikan sebagai model input alternatif. Ini kerana kajian terkini telah menunjukkan penggunaan peta salien dapat meningkatkan prestasi model pembelajaran mendalam dalam pengklasifikasian gambar, pengesanan objek, dan segmentasi gambar. Walau bagaimanapun, sistem konvensional algoritma peta salien jenis bawah-ke-atas kebiasaannya gagal mengesan salien atau anomali patologi dalam gambar-gambar perubatan. Kegagalan ini disebabkan oleh sifat gambar perubatan yang homogen, kurang variasi warna dan kontras. Oleh itu, kajian ini memperkenalkan algoritma Xenafas yang menghasilkan dua jenis pemetaan saliensi anomali iaitu Pemetaan Kebarangkalian Keamatan dan Pemetaan Kebarangkalian Keamatan Pemberat. Kajian dibuat pada peta salien yang dicadangkan iaitu pada lima model pembelajaran mendalam berdasarkan seni bina rangkaian neural konvolusi yang sama. Dapatan kajian menunjukkan dengan menggunakan peta salien atas gambar-gambar radiografi dada tetap membantu kesensitifan kebanyakan model dalam mengidentifikasi gambar-gambar dengan kelegapan ruang udara. Dengan menggunakan algoritma Grad-CAM, peta salien yang dicadangkan ini mampu mengalih fokus model kepada kawasan yang relevan kepada gambar radiografi dada. Sementara itu, kajian kualitatif ini juga menunjukkan algoritma yang dicadangkan mampu memberi ciri anomali, termasuk objek asing dan kardiomegali. Walau bagaimanapun, ianya tidak konsisten dalam menjelaskan berat dan nodul. |
| first_indexed | 2024-12-14T01:18:14Z |
| format | Article |
| id | doaj.art-d4de5407efce467cbf93761d62f0ef0b |
| institution | Directory Open Access Journal |
| issn | 1511-788X 2289-7860 |
| language | English |
| last_indexed | 2024-12-14T01:18:14Z |
| publishDate | 2021-07-01 |
| publisher | IIUM Press, International Islamic University Malaysia |
| record_format | Article |
| series | International Islamic University Malaysia Engineering Journal |
| spelling | doaj.art-d4de5407efce467cbf93761d62f0ef0b2022-12-21T23:22:31ZengIIUM Press, International Islamic University MalaysiaInternational Islamic University Malaysia Engineering Journal1511-788X2289-78602021-07-0122210.31436/iiumej.v22i2.1752CLASSIFICATION OF CHEST RADIOGRAPHS USING NOVEL ANOMALOUS SALIENCY MAP AND DEEP CONVOLUTIONAL NEURAL NETWORKMohd Adli Md Ali0Mohd Radhwan Abidin1Nik Arsyad Nik Muhamad Affendi2Hafidzul Abdullah3Daaniyal R. Rosman4Nu'man Barud'din5Faiz Kemi6Farid Hayati7International Islamic University MalaysiaInternational Islamic University Malaysia International Islamic University Malaysia International Islamic University Malaysia International Islamic University Malaysia International Islamic University Malaysia International Islamic University Malaysia International Islamic University Malaysia The rapid advancement in pattern recognition via the deep learning method has made it possible to develop an autonomous medical image classification system. This system has proven robust and accurate in classifying most pathological features found in a medical image, such as airspace opacity, mass, and broken bone. Conventionally, this system takes routine medical images with minimum pre-processing as the model's input; in this research, we investigate if saliency maps can be an alternative model input. Recent research has shown that saliency maps' application increases deep learning model performance in image classification, object localization, and segmentation. However, conventional bottom-up saliency map algorithms regularly failed to localize salient or pathological anomalies in medical images. This failure is because most medical images are homogenous, lacking color, and contrast variant. Therefore, we also introduce the Xenafas algorithm in this paper. The algorithm creates a new kind of anomalous saliency map called the Intensity Probability Mapping and Weighted Intensity Probability Mapping. We tested the proposed saliency maps on five deep learning models based on common convolutional neural network architecture. The result of this experiment showed that using the proposed saliency map over regular radiograph chest images increases the sensitivity of most models in identifying images with air space opacities. Using the Grad-CAM algorithm, we showed how the proposed saliency map shifted the model attention to the relevant region in chest radiograph images. While in the qualitative study, it was found that the proposed saliency map regularly highlights anomalous features, including foreign objects and cardiomegaly. However, it is inconsistent in highlighting masses and nodules. ABSTRAK: Perkembangan pesat sistem pengecaman corak menggunakan kaedah pembelajaran mendalam membolehkan penghasilan sistem klasifikasi gambar perubatan secara automatik. Sistem ini berupaya menilai secara tepat jika terdapat tanda-tanda patologi di dalam gambar perubatan seperti kelegapan ruang udara, jisim dan tulang patah. Kebiasaannya, sistem ini akan mengambil gambar perubatan dengan pra-pemprosesan minimum sebagai input. Kajian ini adalah tentang potensi peta salien dapat dijadikan sebagai model input alternatif. Ini kerana kajian terkini telah menunjukkan penggunaan peta salien dapat meningkatkan prestasi model pembelajaran mendalam dalam pengklasifikasian gambar, pengesanan objek, dan segmentasi gambar. Walau bagaimanapun, sistem konvensional algoritma peta salien jenis bawah-ke-atas kebiasaannya gagal mengesan salien atau anomali patologi dalam gambar-gambar perubatan. Kegagalan ini disebabkan oleh sifat gambar perubatan yang homogen, kurang variasi warna dan kontras. Oleh itu, kajian ini memperkenalkan algoritma Xenafas yang menghasilkan dua jenis pemetaan saliensi anomali iaitu Pemetaan Kebarangkalian Keamatan dan Pemetaan Kebarangkalian Keamatan Pemberat. Kajian dibuat pada peta salien yang dicadangkan iaitu pada lima model pembelajaran mendalam berdasarkan seni bina rangkaian neural konvolusi yang sama. Dapatan kajian menunjukkan dengan menggunakan peta salien atas gambar-gambar radiografi dada tetap membantu kesensitifan kebanyakan model dalam mengidentifikasi gambar-gambar dengan kelegapan ruang udara. Dengan menggunakan algoritma Grad-CAM, peta salien yang dicadangkan ini mampu mengalih fokus model kepada kawasan yang relevan kepada gambar radiografi dada. Sementara itu, kajian kualitatif ini juga menunjukkan algoritma yang dicadangkan mampu memberi ciri anomali, termasuk objek asing dan kardiomegali. Walau bagaimanapun, ianya tidak konsisten dalam menjelaskan berat dan nodul.https://journals.iium.edu.my/ejournal/index.php/iiumej/article/view/1752Saliency mappingChest RadiographConvolutional Neural Network |
| spellingShingle | Mohd Adli Md Ali Mohd Radhwan Abidin Nik Arsyad Nik Muhamad Affendi Hafidzul Abdullah Daaniyal R. Rosman Nu'man Barud'din Faiz Kemi Farid Hayati CLASSIFICATION OF CHEST RADIOGRAPHS USING NOVEL ANOMALOUS SALIENCY MAP AND DEEP CONVOLUTIONAL NEURAL NETWORK International Islamic University Malaysia Engineering Journal Saliency mapping Chest Radiograph Convolutional Neural Network |
| title | CLASSIFICATION OF CHEST RADIOGRAPHS USING NOVEL ANOMALOUS SALIENCY MAP AND DEEP CONVOLUTIONAL NEURAL NETWORK |
| title_full | CLASSIFICATION OF CHEST RADIOGRAPHS USING NOVEL ANOMALOUS SALIENCY MAP AND DEEP CONVOLUTIONAL NEURAL NETWORK |
| title_fullStr | CLASSIFICATION OF CHEST RADIOGRAPHS USING NOVEL ANOMALOUS SALIENCY MAP AND DEEP CONVOLUTIONAL NEURAL NETWORK |
| title_full_unstemmed | CLASSIFICATION OF CHEST RADIOGRAPHS USING NOVEL ANOMALOUS SALIENCY MAP AND DEEP CONVOLUTIONAL NEURAL NETWORK |
| title_short | CLASSIFICATION OF CHEST RADIOGRAPHS USING NOVEL ANOMALOUS SALIENCY MAP AND DEEP CONVOLUTIONAL NEURAL NETWORK |
| title_sort | classification of chest radiographs using novel anomalous saliency map and deep convolutional neural network |
| topic | Saliency mapping Chest Radiograph Convolutional Neural Network |
| url | https://journals.iium.edu.my/ejournal/index.php/iiumej/article/view/1752 |
| work_keys_str_mv | AT mohdadlimdali classificationofchestradiographsusingnovelanomaloussaliencymapanddeepconvolutionalneuralnetwork AT mohdradhwanabidin classificationofchestradiographsusingnovelanomaloussaliencymapanddeepconvolutionalneuralnetwork AT nikarsyadnikmuhamadaffendi classificationofchestradiographsusingnovelanomaloussaliencymapanddeepconvolutionalneuralnetwork AT hafidzulabdullah classificationofchestradiographsusingnovelanomaloussaliencymapanddeepconvolutionalneuralnetwork AT daaniyalrrosman classificationofchestradiographsusingnovelanomaloussaliencymapanddeepconvolutionalneuralnetwork AT numanbaruddin classificationofchestradiographsusingnovelanomaloussaliencymapanddeepconvolutionalneuralnetwork AT faizkemi classificationofchestradiographsusingnovelanomaloussaliencymapanddeepconvolutionalneuralnetwork AT faridhayati classificationofchestradiographsusingnovelanomaloussaliencymapanddeepconvolutionalneuralnetwork |