| Jul 24 2025
FA EN AR
logo

Advanced Diagnostic & Interventional Radiology Research Center

COVID-19 pandemic 

During the COVID-19 pandemic, the Radiology Research Center at Tehran University of Medical Sciences continued its research activities despite the challenges posed by the increased demand for CT scans of COVID-19 patients and the necessity of adhering to strict health protocols. This center played a crucial role in improving medical imaging techniques, optimizing diagnostic protocols, and advancing technologies related to CT scan image analysis.

Faculty members, researchers, and staff remained committed to ensuring the safety and well-being of healthcare professionals and patients while actively engaging in imaging data analysis, developing artificial intelligence algorithms for faster disease detection, publishing scientific articles, and presenting their findings at international conferences. These efforts aimed to enhance diagnostic accuracy, improve treatment processes, and alleviate pressure on healthcare systems.

 

Key achievements of the Radiology Research Center during the COVID-19 pandemic include:


✔️ Development and optimization of lung imaging protocols for faster and more accurate COVID-19 diagnosis
✔️ Implementation of artificial intelligence technologies for automated CT scan analysis and reduced diagnosis time
✔️ Publication of high-impact research articles on innovative imaging methods for COVID-19 patients
✔️ Participation in national and international projects focused on COVID-19 diagnosis and patient management

The center remains dedicated to advancing research in medical imaging and continues to contribute as a leading scientific institution in improving the quality of diagnostic and therapeutic services.

 

Some of the center's significant achievements during the pandemic include:

 

  • Release Date : Jul 29 2024 - 09:38
  • : 72
  • Study time : 2 minute(s)

A Multi-centric Evaluation of Deep Learning Models for Segmentation of COVID-19 Lung Lesions on Chest CT Scans

{faces}

Background:

 Chest computed tomography (CT) scan is one of the most common tools used for the diagnosis of patients with coronavirus disease 2019 (COVID-19). While segmentation of COVID-19 lung lesions by radiologists can be time-consuming, the application of advanced deep learning techniques for automated segmentation can be a promising step toward the management of this infection and similar diseases in the future.

Objectives:

 This study aimed to evaluate the performance and generalizability of deep learning-based models for the automated segmentation of COVID-19 lung lesions.

Patients and Methods:

 Four datasets (2 private and 2 public) were used in this study. The first and second private datasets included 297 (147 healthy and 150 COVID-19 cases) and 82 COVID-19 subjects. The public datasets included the COVID19-P20 (20 COVID-19 cases from 2 centers) and the MosMedData datasets (50 COVID-19 patients from a single center). Model comparisons were made based on the Dice similarity coefficient (DSC), receiver operating characteristic (ROC) curve, and area under the curve (AUC). The predicted CT severity scores by the model were compared with those of radiologists by measuring the Pearson’s correlation coefficients (PCC). Also, DSC was used to compare the inter-rater agreement of the model and expert against that of 2 experts on an unseen dataset. Finally, the generalizability of the model was evaluated, and a simple calibration strategy was proposed.

Results:

 The VGG16-UNet model showed the best performance across both private datasets, with a DSC of 84.23% ± 1.73% on the first private dataset and 56.61% ± 1.48% on the second private dataset. Similar results were obtained on public datasets, with a DSC of 60.10% ± 2.34% on the COVID19-P20 dataset and 66.28% ± 2.80% on a combined dataset of COVID19-P20 and MosMedData. The predicted CT severity scores of the model were compared against those of radiologists and were found to be 0.89 and 0.85 on the first private dataset and 0.77 and 0.74 on the second private dataset for the right and left lungs, respectively. Moreover, the model trained on the first private dataset was examined on the second private dataset and compared against the radiologist, which revealed a performance gap of 5.74% based on DSCs. A calibration strategy was employed to reduce this gap to 0.53%.

Conclusion:

 The results demonstrated the potential of the proposed model in localizing COVID-19 lesions on CT scans across multiple datasets; its accuracy competed with the radiologists and could assist them in diagnostic and treatment procedures. The effect of model calibration on the performance of an unseen dataset was also reported, increasing the DSC by more than 5%.
  • Article_DOI :
  • Author(s) :
  • News Group : research,publications,articles,covid,covid 19
  • News Code : 277908
مدیر سایت
Author:

مدیر سایت

News list

0 Comments for this article

comment

Post your comment:

متن درون تصویر را در جعبه متن زیر وارد نمائید *

Send to friends

Enter the text inside the security image

Enter the text inside the security image *

clinical centers

useful links

Assessment

  • Online users : 0 person
  • Visitors to this page : 7 person
  • Todays visitors : 36 person
  • Total visitors : 12991 person
  • Last Update : Jul 21 2025 11:39

All site rights belong to Radiology Research Center

niafam espritportal
Enter your desired term to search

Change font color:

Main Color:

Secondary Color:

Change Background Color:

Change font size:

آ-آآ+

Change word space:

Change line space:

Change mouse pointer:

Change font

Choose blind mode

red blindness green blindness blue blindness Red difficult to see greendifficult to see blue difficult to see monochrome nose Monochrome conical nose
Theme settings
Close menu