Drr4covid: Learning Automated COVID-19 Infection Segmentation From Digitally Reconstructed Radiographs

Assessing the Image Quality of Digitally Reconstructed Radiographs from Chest CT

Rising computed tomography (CT) workloads require more efficient image interpretation methods. Digitally reconstructed radiographs (DRRs), generated from CT data, may enhance workflow efficiency by enabling faster radiological assessments. Various techniques exist for generating DRRs. We conducted a pilot study to explore which DRR technique best approaches the image quality of chest X-Rays (CXR). We quantitatively and qualitatively compared four DRR techniques. A retrospective convenience sample of 217 patients who underwent both ultra-low-dose (ULD) chest CT and CXR was used. Four DRRs were generated per ULDCT, and CheXNet, a neural network trained to detect 14 diseases, was applied to CXRs and DRRs to compute area under the curve (AUC) scores. For qualitative assessment, six radiologists rated the image quality of the four DRRs generated from six ULDCTs on a Likert scale from 1 to 6 ('not diagnostic quality' to 'diagnostic quality') and provided feedback, which was analysed using inductive category development. CheXNet's AUC for CXRs was 0.80, while DRR techniques ranged from 0.75 to 0.82 (p > 0.26). Radiologists rated the diagnostic quality of the DRRs between 3.0 and 3.5 on average. The SoftMip technique scored highest in both the quantitative (AUC = 0.82) and the qualitative (score = 3.5) evaluation. DRRs showed comparable disease detection performance to CXRs, suggesting non-inferiority. However, radiologists expressed concerns about DRR image quality, particularly in terms of resolution, noise, and overall look-and-feel. Addressing these limitations with advanced techniques may further align DRRs with the diagnostic standards of CXRs.

Deep learning for automated measurement of CSA related acromion morphological parameters on anteroposterior radiographs

BACKGROUND

The Critical Shoulder Angle Related Acromion Morphological Parameter (CSA- RAMP) is a valuable tool in the analyzing the etiology of the rotator cuff tears (RCTs). However, its clinical application has been limited by the time-consuming and prone to inter- and intra-user variability of the measurement process.

OBJECTIVES

To develop and validate a deep learning algorithm for fully automated assessment of shoulder anteroposterior radiographs associated with RCTs and calculation of CSA-RAMP.

METHODS

Retrospective analysis was conducted on radiographs obtained from computed tomography (CT) scans and X-rays performed between 2018 and 2020 at our institution. The development of the system involved the utilization of digitally reconstructed radiographs (DRRs) generated from each CT scan. The system's performance was evaluated by comparing it with manual and semiautomated measurements on two separate test datasets: dataset I (DRRs) and dataset II (X-rays). Standard metrics, including mean average precision (AP), were utilized to assess the segmentation performance. Additionally, the consistency among fully automated, semiautomated, and manual measurements was comprehensively evaluated using the Pearson correlation coefficient and Bland-Altman analysis.

RESULTS

A total of 1080 DRRs generated from 120 consecutive CT scans and 159 X-ray films were included in the study. The algorithm demonstrated excellent segmentation performance, with a mean AP of 57.67 and an AP50 of 94.31. Strong inter-group correlations were observed for all CSA-RAMP measurements in both test datasets I (automated versus manual, automated versus semiautomated, and semiautomated versus manual; r = [0.990---0.997], P < 0.001) and dataset II (r = [0.984---0.995], P < 0.001). Bland-Altman analysis revealed low bias for all CSA-RAMP measurements in both test datasets I and II, except for CD (with a maximum bias of 2.49%).

CONCLUSIONS

We have successfully developed a fully automated algorithm capable of rapidly and accurately measuring CSA-RAMP on shoulder anteroposterior radiographs. A consistent automated CSA- RAMP measurement system may accelerate powerful and precise studies of disease biology in future large cohorts of RCTs patients.

"Image to patient" equal-resolution surface registration supported by a surface scanner: analysis of algorithm efficiency for computer-aided surgery

PURPOSE

The "image to patient" registration procedure is crucial for the accuracy of surgical instrument tracking relative to the medical image while computer-aided surgery. The main aim of this work was to create an equal-resolution surface registration algorithm (ERSR) and analyze its efficiency.

METHODS

The ERSR algorithm provides two datasets with equal, high resolution and approximately corresponding points. The registered sets are obtained by projection of a user-designed rectangle(s)-shaped uniform clouds of points on DICOM and surface scanner datasets. The tests of the algorithm were performed on a phantom with titanium microscrews. We analyzed the influence of DICOM resolution on the effect of the ERSR algorithm and compared the ERSR to standard paired-points landmark transform registration. The methods of analysis were Target Registration Error, distance maps, and their histogram evaluation.

RESULTS

The mean TRE in case of ERSR equaled 0.8 ± 0.3 mm (resolution A), 0.8 ± 0.5 mm (resolution B), and 1.0 ± 0.7 mm (resolution C). The mean values were at least 0.4 mm lower than in the case of landmark transform registration. The distance maps between the model achieved from the scanner and the CT-based model were analyzed by histogram. The frequency of the first bin in a histogram of the distance map for ERSR was about 0.6 for all three resolutions of DICOM dataset and three times higher than in the case of landmark transform registration. The results were statistically analyzed using the Wilcoxon signed-rank test (alpha = 0.05).

CONCLUSION

The tests proved a statistically significant higher efficiency of equal resolution surface registration related to the landmark transform algorithm. It was proven that the lower resolution of the CT DICOM dataset did not degrade the efficiency of the ERSR algorithm. We observed a significantly lower response to decreased resolution than in the case of paired-points landmark transform registration.