Integrating deep learning algorithms for the lung segmentation and body-part-specific anatomical classification with Medical Imaging and Data Resource Center (MIDRC)

Medical Imaging and Data Resource Center (MIDRC) has been built to support AI-based research in response to the COVID-19 pandemic. One of the main goals of MIDRC is to make data collected in the repository ready for AI analysis. Due to data heterogeneity, there is a need to standardize data and make data-mining easier. Our study aims to stratify imaging data according to underlying anatomy using open-source image processing tools. The experiments were performed using Google Colaboratory on computed tomography (CT) imaging data available from the MIDRC. We adopted the existing open-source tools to process CT series (N=389) to define the image sub-volumes according to body part classification, and additionally identified series slices containing specific anatomic landmarks. Cases with automatically identified chest regions (N=369) were then processed to automatically segment the lungs. In order to assess the accuracy of segmentation, we performed outlier analysis using 3D shape radiomics features extracted from the left and right lungs. Standardized DICOM objects were created to store the resulting segmentations, regions, landmarks and radiomics features. We demonstrated that the MIDRC chest CT collections can be enriched using open-source analysis tools and that data available in MIDRC can be further used to evaluate the robustness of publicly available tools. © 2023 SPIE.

Chest CT in COVID-19 patients: Structured vs conventional reporting.

PURPOSE: To assess clinician satisfaction with structured (SR) and conventional (CR) radiological reports for chest CT exams in coronavirus disease 2019 (COVID-19) patients, objectively comparing both reporting strategies. METHOD: We retrospectively included 68 CTs (61 patients) with COVID-19. CRs were collected from the digital database while corresponding SRs were written by an expert radiologist, including a sign checklist, severity score index and final impressions. New CRs were prepared for a random subset (n = 10) of cases, to allow comparisons in reporting time and word count. CRs were analyzed to record severity score and final impressions inclusion. A random subset of 40 paired CRs and SRs was evaluated by two clinicians to assess, using a Likert scale, readability, comprehensiveness, comprehensibility, conciseness, clinical impact, and overall quality. RESULTS: Overall, 19/68 (28 %) and 9/68 (13 %) of CRs included final impressions and severity score, respectively. SR writing required significantly (p < 0.001) less time (mean = 308 s; SD ± 60 s) compared to CRs (mean = 458 s; SD ± 72 s). On the other hand, word count was not significantly different (p = 0.059, median = 100 and 106, range = 106-139 and 88-131 for SRs and CRs, respectively). Both clinicians expressed significantly (all p < 0.01) higher scores for SRs compared to CRs in all categories. CONCLUSIONS: Our study supports the use of chest CT SRs in COVID-19 patients to improve referring physician satisfaction, optimizing reporting time and provide a greater amount and quality of information within the report.

Coronavirus Disease 2019 (COVID-19) in Italy: Double Reading of Chest CT Examination.

To assess the performance of the second reading of chest compute tomography (CT) examinations by expert radiologists in patients with discordance between the reverse transcription real-time fluorescence polymerase chain reaction (RT-PCR) test for COVID-19 viral pneumonia and the CT report. Three hundred and seventy-eight patients were included in this retrospective study (121 women and 257 men; 71 years median age, with a range of 29-93 years) and subjected to RT-PCR tests for suspicious COVID-19 infection. All patients were subjected to CT examination in order to evaluate the pulmonary disease involvement by COVID-19. CT images were reviewed first by two radiologists who identified COVID-19 typical CT patterns and then reanalyzed by another two radiologists using a CT structured report for COVID-19 diagnosis. Weighted к values were used to evaluate the inter-reader agreement. The median temporal window between RT-PCRs execution and CT scan was zero days with a range of (-9,11) days. The RT-PCR test was positive in 328/378 (86.8%). Discordance between RT-PCR and CT findings for viral pneumonia was revealed in 60 cases. The second reading changed the CT diagnosis in 16/60 (26.7%) cases contributing to an increase the concordance with the RT-PCR. Among these 60 cases, eight were false negative with positive RT-PCR, and 36 were false positive with negative RT-PCR. Sensitivity, specificity, positive predictive value and negative predictive value of CT were respectively of 97.3%, 53.8%, 89.0%, and 88.4%. Double reading of CT scans and expert second readers could increase the diagnostic confidence of radiological interpretation in COVID-19 patients.

Design, validation, and clinical practice of standardized imaging diagnostic report for COVID-19. / COVID-19规范化影像学诊断报告的设计、验证与临床实践.

OBJECTIVES: To design a standardized imaging diagnostic reporting mode for screening coronavirus disease 2019 (COVID-19), and to prospectively verify its effectiveness in clinical practice. METHODS: A new classification and standardized imaging diagnosis report mode of viral pneumonia was established by studying and summarizing the imaging findings of various kinds of viral pneumonia, combining with lesion density, interstitial changes, pleural effusion, lymph nodes, and some special signs. After systematic training, the radiologist experienced clinical practice for screening CT features. COVID-19 cases were screened retrospectively in the single-center. The confirmed cases were verified, and the diagnostic efficacy of the standardized imaging reporting system in screening COVID-19 was tested. RESULTS: There were 912 patients in this stage receiving the screening imaging examination. Of them, 190 patients were screened in the report mode and 30 patients were diagnosed as COVID-19. The CT manifestation of COVID-19 was characterized by pure ground glass lesions or with a few solid components, predominant subpleural distribution, no lymph node enlargement and pleural effusion, and often with paving-way sign and air bronchus sign. In combination with the above signs, the diagnostic efficacy of COVID-19 was 0.942. CONCLUSIONS: The standardized imaging diagnosis report mode based on COVID-19 chest image features is effective and practical, which should be popularized.

Artificial intelligence to codify lung CT in Covid-19 patients.

The spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has already assumed pandemic proportions, affecting over 100 countries in few weeks. A global response is needed to prepare health systems worldwide. Covid-19 can be diagnosed both on chest X-ray and on computed tomography (CT). Asymptomatic patients may also have lung lesions on imaging. CT investigation in patients with suspicion Covid-19 pneumonia involves the use of the high-resolution technique (HRCT). Artificial intelligence (AI) software has been employed to facilitate CT diagnosis. AI software must be useful categorizing the disease into different severities, integrating the structured report, prepared according to subjective considerations, with quantitative, objective assessments of the extent of the lesions. In this communication, we present an example of a good tool for the radiologist (Thoracic VCAR software, GE Healthcare, Italy) in Covid-19 diagnosis (Pan et al. in Radiology, 2020. https://doi.org/10.1148/radiol.2020200370). Thoracic VCAR offers quantitative measurements of the lung involvement. Thoracic VCAR can generate a clear, fast and concise report that communicates vital medical information to referring physicians. In the post-processing phase, software, thanks to the help of a colorimetric map, recognizes the ground glass and differentiates it from consolidation and quantifies them as a percentage with respect to the healthy parenchyma. AI software therefore allows to accurately calculate the volume of each of these areas. Therefore, keeping in mind that CT has high diagnostic sensitivity in identifying lesions, but not specific for Covid-19 and similar to other infectious viral diseases, it is mandatory to have an AI software that expresses objective evaluations of the percentage of ventilated lung parenchyma compared to the affected one.