PMC:7782580 / 11165-13684 JSONTXT

{"project":"LitCovid-PubTator","denotations":[{"id":"127","span":{"begin":55,"end":60},"obj":"Species"},{"id":"128","span":{"begin":83,"end":88},"obj":"Species"},{"id":"129","span":{"begin":1376,"end":1384},"obj":"Disease"}],"attributes":[{"id":"A127","pred":"tao:has_database_id","subj":"127","obj":"Tax:9606"},{"id":"A128","pred":"tao:has_database_id","subj":"128","obj":"Tax:9606"},{"id":"A129","pred":"tao:has_database_id","subj":"129","obj":"MESH:C000657245"}],"namespaces":[{"prefix":"Tax","uri":"https://www.ncbi.nlm.nih.gov/taxonomy/"},{"prefix":"MESH","uri":"https://id.nlm.nih.gov/mesh/"},{"prefix":"Gene","uri":"https://www.ncbi.nlm.nih.gov/gene/"},{"prefix":"CVCL","uri":"https://web.expasy.org/cellosaurus/CVCL_"}],"text":"Medical imaging uses images of internal tissues of the human body or a part of the human body in a non-invasive manner for clinical diagnoses or treatment plans36. Medical images (e.g., X-data and CT-data) are usually acquired using computed radiography and are typically stored in the Digital Imaging and Communications in Medicine (DICOM) format37. X-data are two-dimensional grayscale images, and CT-data are three-dimensional data, consisting of slices of the data in the z axis direction of a two-dimensional grayscale image. Machine learning methods are playing increasingly important roles in medical image analysis, especially DL methods. DL uses multiple non-linear transformations to create a mapping relationship between the input data and output labels38. The objective of this study was to annotate lesion areas in medical images with high accuracy. Therefore, we developed a pseudo-coloring method, which is a technique that helps enhance medical images for physicians to isolate relevant tissues and groups different tissues together39. We converted the original grayscale images to color images using the open-source image processing tools Open Source Computer Vision Library (OpenCV) and Pillow. Examples of the pseudo-color images are shown in Fig. 1a. We developed a platform that uses a client-server architecture to annotate the potential lesion areas of COVID-19 on the CXR and CT images. The platform can be deployed on a private cloud for security and local sharing. All the images were annotated by two experienced radiologists (one was a 5th-year radiologist and the other was a 3rd-year radiologist) in the Youan Hospital. If there was disagreement about a result, a senior radiologist and a respiratory doctor made the final decision to ensure the precision of the annotation process. The details of the annotation pipeline are shown in Supplementary Fig. 1.\nFig. 1 Demonstrations of data preprocessing methods including pseudo-coloring and dimension normalization.\na Pseudo-coloring for abnormal examples in the CXR and CT images. The original grayscale images were transformed into color images using the pseudo-coloring method and were annotated by the experts. The scale bar on the right is the range of pixel values of the image data. b Dimension normalization to reduce the dimensions in the CT images. The number of CT images were first resampled to a multiple of three and then divided into three groups. Followed by the 1 × 1 convolution layers to reduce the dimensions of the data."}

{"project":"LitCovid-sentences","denotations":[{"id":"T81","span":{"begin":0,"end":163},"obj":"Sentence"},{"id":"T82","span":{"begin":164,"end":350},"obj":"Sentence"},{"id":"T83","span":{"begin":351,"end":530},"obj":"Sentence"},{"id":"T84","span":{"begin":531,"end":646},"obj":"Sentence"},{"id":"T85","span":{"begin":647,"end":767},"obj":"Sentence"},{"id":"T86","span":{"begin":768,"end":862},"obj":"Sentence"},{"id":"T87","span":{"begin":863,"end":1051},"obj":"Sentence"},{"id":"T88","span":{"begin":1052,"end":1212},"obj":"Sentence"},{"id":"T89","span":{"begin":1213,"end":1270},"obj":"Sentence"},{"id":"T90","span":{"begin":1271,"end":1410},"obj":"Sentence"},{"id":"T91","span":{"begin":1411,"end":1490},"obj":"Sentence"},{"id":"T92","span":{"begin":1491,"end":1649},"obj":"Sentence"},{"id":"T93","span":{"begin":1650,"end":1812},"obj":"Sentence"},{"id":"T94","span":{"begin":1813,"end":1886},"obj":"Sentence"},{"id":"T95","span":{"begin":1887,"end":1993},"obj":"Sentence"},{"id":"T96","span":{"begin":1994,"end":2059},"obj":"Sentence"},{"id":"T97","span":{"begin":2060,"end":2192},"obj":"Sentence"},{"id":"T98","span":{"begin":2193,"end":2336},"obj":"Sentence"},{"id":"T99","span":{"begin":2337,"end":2440},"obj":"Sentence"},{"id":"T100","span":{"begin":2441,"end":2519},"obj":"Sentence"}],"namespaces":[{"prefix":"_base","uri":"http://pubannotation.org/ontology/tao.owl#"}],"text":"Medical imaging uses images of internal tissues of the human body or a part of the human body in a non-invasive manner for clinical diagnoses or treatment plans36. Medical images (e.g., X-data and CT-data) are usually acquired using computed radiography and are typically stored in the Digital Imaging and Communications in Medicine (DICOM) format37. X-data are two-dimensional grayscale images, and CT-data are three-dimensional data, consisting of slices of the data in the z axis direction of a two-dimensional grayscale image. Machine learning methods are playing increasingly important roles in medical image analysis, especially DL methods. DL uses multiple non-linear transformations to create a mapping relationship between the input data and output labels38. The objective of this study was to annotate lesion areas in medical images with high accuracy. Therefore, we developed a pseudo-coloring method, which is a technique that helps enhance medical images for physicians to isolate relevant tissues and groups different tissues together39. We converted the original grayscale images to color images using the open-source image processing tools Open Source Computer Vision Library (OpenCV) and Pillow. Examples of the pseudo-color images are shown in Fig. 1a. We developed a platform that uses a client-server architecture to annotate the potential lesion areas of COVID-19 on the CXR and CT images. The platform can be deployed on a private cloud for security and local sharing. All the images were annotated by two experienced radiologists (one was a 5th-year radiologist and the other was a 3rd-year radiologist) in the Youan Hospital. If there was disagreement about a result, a senior radiologist and a respiratory doctor made the final decision to ensure the precision of the annotation process. The details of the annotation pipeline are shown in Supplementary Fig. 1.\nFig. 1 Demonstrations of data preprocessing methods including pseudo-coloring and dimension normalization.\na Pseudo-coloring for abnormal examples in the CXR and CT images. The original grayscale images were transformed into color images using the pseudo-coloring method and were annotated by the experts. The scale bar on the right is the range of pixel values of the image data. b Dimension normalization to reduce the dimensions in the CT images. The number of CT images were first resampled to a multiple of three and then divided into three groups. Followed by the 1 × 1 convolution layers to reduce the dimensions of the data."}