Journal of Medical Imaging最新文献

Purpose: We evaluate the impact of charge summing correction on a cadmium telluride (CdTe)-based photon-counting detector in breast computed tomography (CT).

Approach: We employ a custom-built laboratory benchtop system using the X-THOR FX30 0.75-mm CdTe detector (Varex Imaging, Salt Lake City, Utah, United States) with a pixel pitch of 0.1 mm, operated in both standard mode [single pixel (SP)] and charge summing correction mode [anticoincidence (AC)]. A tungsten anode source operated at 55 kVp with 2-mm aluminum external filtration and tube currents of 25, 100, and 200 mA with corresponding exposure times of 20, 5, and 2.5 ms were employed to study the effects of X-ray fluence and pulse pileup. Performance comparisons between AC and SP modes are performed in both projection and image reconstructed spaces. In the projection space, performance metrics include count rate, energy resolution, uniformity, modulation transfer function (MTF), and noise power spectrum (NPS). In the image space, performance metrics consist of contrast-to-noise ratio (CNR), uniformity, NPS, and iodine quantification accuracy. For both acquisition modes, signal-to-thickness calibration, for gain and beam hardening corrections, is used before image reconstruction. Images are reconstructed via TIGRE CT software using the standard Feldkamp, Davis, and Kress (FDK) filtered back projection algorithm with a Hann filter and reconstructed with a voxel size of 0.081 mm. Material decomposition is performed using a standard image-based method.

Results: In the detector space, the application of hardware-based charge summing correction enhances spectral resolution and improves the spatial resolution of MTF at lower energy thresholds but introduces anomalous edge enhancement effects and artifacts in the MTF at high fluence. A negative noise correlation was observed in AC mode-acquired images. As expected, the AC acquisition mode results in a decreased detector count rate. In the image space, NPS results displayed elevated noise in low-energy AC images. However, at high energy, noise was comparable between both modes. Greater uniformity was observed in SP mode-acquired images. The largest disparity was observed in the iodine quantification test, where the AC mode demonstrates a much stronger linear relationship between estimated and true iodine concentrations than the SP mode.

Conclusion: The results are specific to the studied system, reconstruction parameters, and irradiation conditions limited to 200 mA and 0.5 mAs. The AC mode generally provides better energy and MTF resolution at low energy thresholds but with increased noise and reduced uniformity. In image space, charge summing correction improved iodine quantification and CNR at high energy thresholds.

Purpose: We aim to investigate the characteristics and evaluate the performance of synthetic mammograms (SMs) based on wide-angle digital breast tomosynthesis (DBT) compared with digital mammography (DM).

Approach: Fifty cases with both synthetic and digital mammograms were selected from the Malmö Breast Tomosynthesis Screening Trial. They were categorized into five groups consisting of normal cases and recalled cases with false-positive and true-positive findings from DM and DBT only. The DBT system used was a wide-angle (WA) system from Siemens, and the SM images were reconstructed from the DBT images. Visual grading, detection, and recall were evaluated by experienced breast radiologists in both SM and DM images.

Results: Some image quality criteria of the SM images were rated as qualitatively inferior to DM images. However, reader-averaged diagnostic accuracy (0.57 versus 0.55), sensitivity (0.46 versus 0.50), and specificity (0.64 versus 0.58) were not significantly different between SM and DM, respectively.

Conclusions: Synthetic mammography plays a promising role to complement or even replace DM. The study could not find any indications of substantial differences in the sensitivity or specificity of SM for WA DBT systems compared with DM. However, certain image quality criteria of SM fall slightly short compared with DM images. Next-generation DBT systems could address such limitations through improved reconstruction algorithms and system design, and their performance should be the focus of future research studies.

Purpose: Wireless capsule endoscopy (WCE) is a non-invasive technology used for diagnosing gastrointestinal abnormalities. A single examination generates $\sim \mathrm{55,000}$ images, making manual review both time-consuming and costly for doctors. Therefore, the development of computer vision-assisted systems is highly desirable to aid in the diagnostic process.

Approach: We presents a deep learning approach leveraging knowledge distillation (KD) from a convolutional neural network (CNN) teacher model to a vision transformer (ViT) student model for gastrointestinal abnormality recognition. The CNN teacher model utilizes attention mechanisms and depth-wise separable convolutions to extract features from WCE images, supervising the ViT in learning these representations.

Results: The proposed method achieves accuracy of 97% and 96% on the Kvasir and KID datasets, respectively, demonstrating its effectiveness in distinguishing normal from abnormal regions and bleeding from non-bleeding cases. The proposed approach offers computational efficiency and generalization to unseen datasets, outperforming several state-of-the-art methods.

Conclusions: We proposed a deep learning approach utilizing CNNs and a ViT with KD to effectively classify gastrointestinal diseases in WCE images. It demonstrates promising performance on public datasets, distinguishing normal from abnormal regions and bleeding from non-bleeding cases while offering optimal computational efficiency compared with existing methods, making it suitable for GI disease applications.

Purpose: We aim to investigate the localization, visibility, and measurement of lung nodules in digital chest tomosynthesis (DTS).

Approach: Computed tomography (CT), maximum intensity projections (CT-MIP) (transaxial versus coronal orientation), and computer-aided detection (CAD) were used as location reference, and inter- and intra-observer agreement regarding lung nodule size was assessed. Five radiologists analyzed DTS and CT images from 24 participants with lung $\text{nodules}\ge 100{\mathrm{mm}}^3$ , focusing on lung nodule localization, visibility, and measurement on DTS. Visual grading was used to compare if coronal or transaxial CT-MIP better facilitated the localization of lung nodules in DTS.

Results: The majority of the lung nodules (79%) were rated as visible in DTS, although less clearly in comparison with CT. Coronal CT-MIP was the preferred orientation in the task of locating nodules on DTS. On DTS, area-based lung nodule size estimates resulted in significantly less measurement variability when compared with nodule size estimated based on mean diameter (mD) ( $p<0.05$ ). Also, on DTS, area-based lung nodule size estimates were more accurate ( $\mathrm{SEE}=38.7{\mathrm{mm}}^3$ ) than lung nodule size estimates based on mean diameter ( $\mathrm{SEE}=42.7{\mathrm{mm}}^3$ ).

Conclusions: Coronal CT-MIP images are superior to transaxial CT-MIP images in facilitating lung nodule localization in DTS. Most $\text{nodules}\ge 100{\mathrm{mm}}^3$ found on CT can be visualized, correctly localized, and measured in DTS, and area-based measurement may be the key to more precise and less variable nodule measurements on DTS.

Purpose: Chest tomosynthesis (CTS) has a relatively longer acquisition time compared with chest X-ray, which may increase the risk of motion artifacts in the reconstructed images. Motion artifacts induced by breathing motion adversely impact the image quality. This study aims to reduce these artifacts by excluding projection images identified with breathing motion prior to the reconstruction of section images and to assess if motion compensation improves overall image quality.

Approach: In this study, 2969 CTS examinations were analyzed to identify examinations where breathing motion has occurred using a method based on localizing the diaphragm border in each of the projection images. A trajectory over diaphragm positions was estimated from a second-order polynomial curve fit, and projection images where the diaphragm border deviated from the trajectory were removed before reconstruction. The image quality between motion-compensated and uncompensated examinations was evaluated using the image quality criteria for anatomical structures and image artifacts in a visual grading characteristic (VGC) study. The resulting rating data were statistically analyzed using the software VGC analyzer.

Results: A total of 58 examinations were included in this study with breathing motion occurring either at the beginning or end ( $n=17$ ) or throughout the entire acquisition ( $n=41$ ). In general, no significant difference in image quality or presence of motion artifacts was shown between the motion-compensated and uncompensated examinations. However, motion compensation significantly improved the image quality and reduced the motion artifacts in cases where motion occurred at the beginning or end. In examinations where motion occurred throughout the acquisition, motion compensation led to a significant increase in ripple artifacts and noise.

Conclusions: Compensation for respiratory motion in CTS by excluding projection images may improve the image quality if the motion occurs mainly at the beginning or end of the examination. However, the disadvantages of excluding projections may outweigh the benefits of motion compensation.

Purpose: The color of Papanicolaou-stained specimens is a crucial feature in cytology diagnosis. However, the quantification of color using digital images is challenging due to the variations in the staining process and characteristics of imaging equipment. The dye amount estimation of stained specimens is helpful for quantitatively interpreting the color based on a physical model. It has been realized with color unmixing and applied to staining with three or fewer dyes. Nevertheless, the Papanicolaou stain comprises five dyes. Thus, we employ multispectral imaging with more channels for quantitative analysis of the Papanicolaou-stained cervical cytology samples.

Approach: We estimate the dye amount map from a 14-band multispectral observation capturing a Papanicolaou-stained specimen using the actual measured spectral characteristics of the single-stained samples. The estimated dye amount maps were employed for the quantitative interpretation of the color of cytoplasmic mucin of lobular endocervical glandular hyperplasia (LEGH) and normal endocervical (EC) cells in a uterine cervical lesion.

Results: We demonstrated the dye amount estimation performance of the proposed method using single-stain images and Papanicolaou-stain images. Moreover, the yellowish color in the LEGH cells is found to be interpreted with more orange G (OG) and less Eosin Y (EY) dye amounts. We also elucidated that LEGH and EC cells could be classified using linear classifiers from the dye amount.

Conclusions: Multispectral imaging enables the quantitative analysis of dye amount maps of Papanicolaou-stained cytology specimens. The effectiveness is demonstrated in interpreting and classifying the cytoplasmic mucin of EC and LEGH cells in cervical cytology.

Purpose: Differentiating primary central nervous system lymphoma (PCNSL) and glioblastoma (GBM) is crucial because their prognosis and treatment differ substantially. Manual examination of their histological characteristics is considered the golden standard in clinical diagnosis. However, this process is tedious and time-consuming and might lead to misdiagnosis caused by morphological similarity between their histology and tumor heterogeneity. Existing research focuses on radiological differentiation, which mostly uses multi-parametric magnetic resonance imaging. By contrast, we investigate the pathological differentiation between the two types of tumors using whole slide images (WSIs) of postoperative formalin-fixed paraffin-embedded samples.

Approach: To learn the specific and intrinsic histological feature representations from the WSI patches, a self-supervised feature extractor is trained. Then, the patch representations are fused by feeding into a weakly supervised multiple-instance learning model for the WSI classification. We validate our approach on 134 PCNSL and 526 GBM cases collected from three hospitals. We also investigate the effect of feature extraction on the final prediction by comparing the performance of applying the feature extractors trained on the PCNSL/GBM slides from specific institutions, multi-site PCNSL/GBM slides, and large-scale histopathological images.

Results: Different feature extractors perform comparably with the overall area under the receiver operating characteristic curve value exceeding 85% for each dataset and close to 95% for the combined multi-site dataset. Using the institution-specific feature extractors generally obtains the best overall prediction with both of the PCNSL and GBM classification accuracies reaching 80% for each dataset.

Conclusions: The excellent classification performance suggests that our approach can be used as an assistant tool to reduce the pathologists' workload by providing an accurate and objective second diagnosis. Moreover, the discriminant regions indicated by the generated attention heatmap improve the model interpretability and provide additional diagnostic information.

Purpose: The survival rate of breast cancer for women in low- and middle-income countries is poor compared with that in high-income countries. Point-of-care ultrasound (POCUS) combined with deep learning could potentially be a suitable solution enabling early detection of breast cancer. We aim to improve a classification network dedicated to classifying POCUS images by comparing different techniques for increasing the amount of training data.

Approach: Two data sets consisting of breast tissue images were collected, one captured with POCUS and another with standard ultrasound (US). The data sets were expanded by using different techniques, including augmentation, histogram matching, histogram equalization, and cycle-consistent adversarial networks (CycleGANs). A classification network was trained on different combinations of the original and expanded data sets. Different types of augmentation were investigated and two different CycleGAN approaches were implemented.

Results: Almost all methods for expanding the data sets significantly improved the classification results compared with solely using POCUS images during the training of the classification network. When training the classification network on POCUS and CycleGAN-generated POCUS images, it was possible to achieve an area under the receiver operating characteristic curve of 95.3% (95% confidence interval 93.4% to 97.0%).

Conclusions: Applying augmentation during training showed to be important and increased the performance of the classification network. Adding more data also increased the performance, but using standard US images or CycleGAN-generated POCUS images gave similar results.

Purpose: Contrast-enhanced digital breast tomosynthesis (CEDBT) highlights breast tumors with neo-angiogenesis. A recently proposed CEDBT system with a dual-layer (DL) flat-panel detector enables simultaneous acquisition of high-energy (HE) and low-energy (LE) projection images with a single exposure, which reduces acquisition time and eliminates motion artifacts. However, x-ray scatter degrades image quality and lesion detectability. We propose a practical method for accurate and robust scatter correction (SC) for DL-CEDBT.

Approach: The proposed hybrid SC method combines the advantages of a two-kernel iterative convolution method and an empirical interpolation strategy, which accounts for the reduced scatter from the peripheral breast region due to thickness roll-off and the scatter contribution from the region outside the breast. Scatter point spread functions were generated using Monte Carlo simulations with different breast glandular fractions, compressed thicknesses, and projection angles. Projection images and ground truth scatter maps of anthropomorphic digital breast phantoms were simulated to evaluate the performance of the proposed SC method and three other kernel- and interpolation-based methods. The mean absolute relative error (MARE) between scatter estimates and ground truth was used as the metric for SC accuracy.

Results: DL-CEDBT shows scatter characteristics different from dual-shot, primarily due to the two energy peaks of the incident spectrum and the structure of the DL detector. Compared with the other methods investigated, the proposed hybrid SC method showed superior accuracy and robustness, with MARE of $\sim 3.1\%$ for all LE and HE projection images of different phantoms in both cranial-caudal and mediolateral-oblique views. After SC, cupping artifacts in the dual-energy image were removed, and the signal difference-to-noise ratio was improved by 82.0% for 8 mm iodine objects.

Conclusions: A practical SC method was developed, which provided accurate and robust scatter estimates to improve image quality and lesion detectability for DL-CEDBT.

Purpose: The purpose of this study was to develop and evaluate an algorithm for calculating a shape feature to automatically detect both main pancreatic duct dilation (MPDD) and pancreatic parenchymal atrophy (PPA) in abdominal contrast-enhanced CT (CE-CT) images.

Approach: The proposed algorithm for the automatic detection of MPDD and PPA is composed of five processes: coarse pancreas segmentation, fine pancreas segmentation, main pancreatic duct (MPD) segmentation, centerline estimation, and shape feature calculation. First, the pancreas region is segmented by a deep learning convolutional neural network (CNN). Then, the MPD region is segmented inside the pancreatic region by the deep learning CNN. Next, centerline estimation is performed using Dijkstra's rooting algorithm. Finally, in shape feature calculation, the cross-sectional area ratio of the pancreatic duct to the pancreatic parenchyma (DP ratio) is calculated in all cross sections perpendicular to the identified centerline, and the 90th percentile value of the DP ratio for all cross sections (90th DP ratio) is calculated. The detection performance of the 90th DP ratio for MPDD and PPA was evaluated using 56 abdominal CE-CT images available as public data.

Results: The average of the 90th DP ratio was 0.059 in 48 cases with MPDD and 0.007 in eight cases without MPDD ( $p<0.001$ ) and 0.074 in 31 cases with PPA and 0.023 in 25 cases without PPA ( $p<0.001$ ).

Conclusions: We have developed an algorithm for calculating an automatically measurable shape feature called the 90th DP ratio for the detection of MPDD and PPA.