Quantitative Imaging in Medicine and Surgery最新文献

Pulmonary artery aneurysm (PAA) is a rare pulmonary vascular disease with nonspecific symptoms and various etiologies. As the disease progresses, in addition to the dilation of the pulmonary arteries, it may be accompanied by remodeling of the cardiac structure and changes in the morphology of the aorta. Recognizing the cause of PAA is therefore a clinically challenging task. In this review article, we provide an overview of various causes of PAA with the support of corresponding imaging findings on computed tomography pulmonary angiography (CTPA) examination. Firstly, from the perspective of hemodynamics, a logical diagnosis is provided according to whether the main pulmonary artery (MPA) is dilated, and whether the PA is dilated locally or diffusely. Secondly, for the imaging examination of vascular wall lesions, due to the limitations of ultrasound examination and interventional procedures, the irreplaceability of dual-phase CTPA examination in disease assessment is especially emphasized. Finally, for highly suspected disorders, it is necessary to comprehensively check with the patient whether there is a family history or past medical history. For patients with PAA, especially those with Marfan syndrome (MFS) or arteritis, adequate preoperative imaging evaluation, regular postoperative radiographic follow-up, and concurrent treatment of the underlying disease (if necessary) are crucial, which are related to the prognosis and long-term quality of life of such patients. Despite the nonspecific features of PAA presentation, a thorough examination of the patient's clinical history and imaging characteristics will play an important role in diagnosing PAA and planning patient management strategies.

Background: Meniere's disease (MD) is an ear-related vestibular disorder accompanied by vertigo, hearing loss, and tinnitus. The anatomical structure and spatial position of the semicircular canals are important for understanding vestibular function and disease; however, research on MD and the effect of anatomical changes in the semicircular canals is limited. This study explored the relationship between the spatial location of the semicircular canals and MD using ultra-high-resolution computed tomography (U-HRCT) and intelligent segmentation.

Methods: Isotropic U-HRCT images obtained from patients with MD and healthy controls (HCs) were retrospectively analyzed. We extracted the semicircular canal structures and extracted their skeleton. The plane of the skeleton of each semicircular canal was fitted separately. The mutual angles between the semicircular canals, and the angles between each semicircular canal and each plane of the coordinate system were measured.

Results: Among 45 MD-affected ears (MDAEs), 33 MD-healthy ears (MDHEs), and 45 HC ears, the angle between the superior and lateral semicircular canals (LSCs) and the angle between the superior and posterior semicircular canals (PSCs) were larger in the MDAE and MDHE groups than the HC group (P<0.01), while the angle between the posterior and LSCs was smaller in the MDAE group than the HC group (P<0.001). The angles between the superior and PSCs and coronal plane (CP) of the coordinate system were significantly smaller in the MDAE and MDHE groups than the HC group (P<0.01); however, the angles between the LSC and axial plane and CP were significantly larger in the MDAE and MDHE groups than the HC group (P<0.001).

Conclusions: Spatial position changes in the semicircular canals may be the anatomical basis of MD.

Background: Most primary bone tumors are often found in the bone around the knee joint. However, the detection of primary bone tumors on radiographs can be challenging for the inexperienced or junior radiologist. This study aimed to develop a deep learning (DL) model for the detection of primary bone tumors around the knee joint on radiographs.

Methods: From four tertiary referral centers, we recruited 687 patients diagnosed with bone tumors (including osteosarcoma, chondrosarcoma, giant cell tumor of bone, bone cyst, enchondroma, fibrous dysplasia, etc.; 417 males, 270 females; mean age 22.8±13.2 years) by postoperative pathology or clinical imaging/follow-up, and 1,988 participants with normal bone radiographs (1,152 males, 836 females; mean age 27.9±12.2 years). The dataset was split into a training set for model development, an internal independent and an external test set for model validation. The trained model located bone tumor lesions and then detected tumor patients. Receiver operating characteristic curves and Cohen's kappa coefficient were used for evaluating detection performance. We compared the model's detection performance with that of two junior radiologists in the internal test set using permutation tests.

Results: The DL model correctly localized 94.5% and 92.9% bone tumors on radiographs in the internal and external test set, respectively. An accuracy of 0.964/0.920, and an area under the receiver operating characteristic curve (AUC) of 0.981/0.990 in DL detection of bone tumor patients were for the internal and external test set, respectively. Cohen's kappa coefficient of the model in the internal test set was significantly higher than that of the two junior radiologists with 4 and 3 years of experience in musculoskeletal radiology (Model vs. Reader A, 0.927 vs. 0.777, P<0.001; Model vs. Reader B, 0.927 vs. 0.841, P=0.033).

Conclusions: The DL model achieved good performance in detecting primary bone tumors around the knee joint. This model had better performance than those of junior radiologists, indicating the potential for the detection of bone tumors on radiographs.

Background: Dynamic chest radiography (DCR) is a novel and supplementary examination in respiratory diseases. The investigation of other chest diseases using DCR has been explored, identifying a certain correlation of the pulmonary function test (PFT). However, there is a lack of research using DCR parameters to quantitatively evaluate chest disease. The purpose of this study was to investigate the diagnostic value of DCR for diaphragm paralysis (DP).

Methods: This retrospective study recruited 118 participants, which include 18 patients with DP, 48 healthy volunteers, and 52 patients with respiratory disease. Comparison of DCR parameters relationships among 3 groups was performed using one-way analysis of variance (ANOVA) and Kruskal-Wallis test. The receiver operating characteristic (ROC) curve was used to compare the value of the DCR parameters to diagnose DP.

Results: The differences of excursion of diaphragm (ED) in normal (nb) and forced breathing (fb), ED(fb)-ED(nb), and the parameters of projected lung area (PLA) in inspiratory (ins) and expiratory phase (exp), PLA.exp(fb), PLA.ins(fb)-PLA.ins(nb), and PLA.exp(fb)-PLA.exp(nb) among the 3 groups were statistically significant. The highest area under the curve (AUC) of right-side parameter was the ED(fb)-ED(nb), for which the AUC was 0.8950 [95% confidence interval (CI): 0.7618-1.000], whereas that of the left-side parameter was ED(fb), for which the AUC was 0.9176 [95% confidence interval (CI): 0.8524-0.9829].

Conclusions: The parameters of DCR have good diagnostic value for DP. The highest diagnostic efficiency for DP on the right side is the ED(fb)-ED(nb), with a sensitivity of 95% and a specificity of 78.6%, whereas on the left side is ED(fb), with a sensitivity of 80% and a specificity of 88.2%.

Background: Axial spondyloarthritis (axSpA) is frequently diagnosed late, particularly in human leukocyte antigen (HLA)-B27-negative patients, resulting in a missed opportunity for optimal treatment. This study aimed to develop an artificial intelligence (AI) tool, termed NegSpA-AI, using sacroiliac joint (SIJ) magnetic resonance imaging (MRI) and clinical SpA features to improve the diagnosis of axSpA in HLA-B27-negative patients.

Methods: We retrospectively included 454 HLA-B27-negative patients with rheumatologist-diagnosed axSpA or other diseases (non-axSpA) from the Third Affiliated Hospital of Southern Medical University and Nanhai Hospital between January 2010 and August 2021. They were divided into a training set (n=328) for 5-fold cross-validation, an internal test set (n=72), and an independent external test set (n=54). To construct a prospective test set, we further enrolled 87 patients between September 2021 and August 2023 from the Third Affiliated Hospital of Southern Medical University. MRI techniques employed included T1-weighted (T1W), T2-weighted (T2W), and fat-suppressed (FS) sequences. We developed NegSpA-AI using a deep learning (DL) network to differentiate between axSpA and non-axSpA at admission. Furthermore, we conducted a reader study involving 4 radiologists and 2 rheumatologists to evaluate and compare the performance of independent and AI-assisted clinicians.

Results: NegSpA-AI demonstrated superior performance compared to the independent junior rheumatologist (≤5 years of experience), achieving areas under the curve (AUCs) of 0.878 [95% confidence interval (CI): 0.786-0.971], 0.870 (95% CI: 0.771-0.970), and 0.815 (95% CI: 0.714-0.915) on the internal, external, and prospective test sets, respectively. The assistance of NegSpA-AI promoted discriminating accuracy, sensitivity, and specificity of independent junior radiologists by 7.4-11.5%, 1.0-13.3%, and 7.4-20.6% across the 3 test sets (all P<0.05). On the prospective test set, AI assistance also improved the diagnostic accuracy, sensitivity, and specificity of independent junior rheumatologists by 7.7%, 7.7%, and 6.9%, respectively (all P<0.01).

Conclusions: The proposed NegSpA-AI effectively improves radiologists' interpretations of SIJ MRI and rheumatologists' diagnoses of HLA-B27-negative axSpA.

Background: An understanding of the anatomical structure is crucial for completing successful endoscopic dacryocystorhinostomy (DCR) surgery. This study aimed to precisely delineate the spatial relationship between the lacrimal sac and the agger nasi cell (ANC) and evaluate the impact of ANC on surgical strategies in endoscopic DCR.

Methods: This retrospective cross-sectional study included 110 Han Chinese patients diagnosed with unilateral primary acquired nasolacrimal duct obstruction (PANDO) from January 2021 to June 2023. This study was conducted in Eye, Ear, Nose, and Throat Hospital of Fudan University and involved inpatient participants who were scheduled for DCR surgery under general anesthesia. Patients were consecutively enrolled. The patients underwent preoperative computed tomography-dacryocystography (CT-DCG), and contrast-enhanced images were used to locate the positions of the lacrimal sac and the common canaliculus. A dynamic approach was adopted to analyze the multiplanar CT imaging, facilitating a detailed assessment of the morphology of the lacrimal drainage system and potential overlap of the lacrimal sac. Patient ages and measured values are presented as the mean ± standard deviation, which were measured three times by the same observer and averaged for statistical analysis.

Results: The prevalence of ANC in this study was 90.9% (100/110). Dynamic examination revealed that only 42.7% (47/110) of ANCs appeared as discrete cells, while the majority were connected to nearby sinus openings. Spatial analysis showed that in 57 out of 110 cases, ANCs were situated below the common canaliculus and not posterior to the lacrimal sac, indicating an overlap rate of 51.8%. Notably, our dynamic approach identified five critical cases of overlap below the level of the common canaliculus, which might have been missed by prior studies that used different methodologies.

Conclusions: More than half of the ANCs exhibited overlap with the lacrimal sac, suggesting a significant proportion may necessitate opening during endoscopic DCR procedures. ANCs are often interconnected with adjacent nasal sinuses, necessitating careful consideration in the decision to open the ANCs during surgery. The dynamic evaluation employed in CT-DCG effectively assessed the extent of ANC coverage over the lacrimal sac.

Background: Morphological parameters of the lumbar spine are valuable in assessing lumbar spine diseases. However, manual measurement of lumbar morphological parameters is time-consuming. Deep learning has automatic quantitative and qualitative analysis capabilities. To develop a deep learning-based model for the automatic quantitative measurement of morphological parameters from anteroposterior digital radiographs of the lumbar spine and to evaluate its performance.

Methods: This study used 1,368 anteroposterior digital radiographs of the lumbar spine to train a deep learning model to measure the quantitative morphological indicators, including L1 to L5 vertebral body height (VBH) and L1-L2 to L4-L5 intervertebral disc height (IDH). The means of the manual measurements by three radiologists were used as the reference standard. The parameters predicted by the model were analyzed against the manual measurements using paired t-tests. Percentage of correct key points (PCK), intra-class correlation coefficient (ICC), Pearson correlation coefficient (r), mean absolute error (MAE), root mean square error (RMSE), and Bland-Altman plots were performed to assess the performance of the model.

Results: Within the 3-mm distance threshold, the model had a PCK range of 99.77-99.46% for the L1 to L4 vertebrae and 77.37% for the L5 vertebrae. Except for VBH-L5 and IDH_L3-L4, IDH_L4-L5 (P<0.05), the estimated values of the model in the remaining parameters were not statistically significant compared with the reference standard (P>0.05). Except for VBH-L5 and IDH_L4-L5, the model showed good correlation and consistency with the reference standard (ICC =0.84-0.96, r=0.85-0.97, MAE =0.5-0.66, RMSE =0.66-0.95). The model outperformed other models (EfficientDet + Unet, EfficientDet + DarkPose, HRNet, and Unet) in predicting landmarks within a distance threshold of 1.5 to 5 mm.

Conclusions: The model developed in this study can automatically measure the morphological parameters of the L1 to L4 vertebrae from anteroposterior digital radiographs of the lumbar spine. Its performance is close to the level of radiologists.

Background: Bone age assessment (BAA) is crucial for the diagnosis of growth disorders and the optimization of treatments. However, the random error caused by different observers' experiences and the low consistency of repeated assessments harms the quality of such assessments. Thus, automated assessment methods are needed.

Methods: Previous research has sought to design localization modules in a strongly or weakly supervised fashion to aggregate part regions to better recognize subtle differences. Conversely, we sought to efficiently deliver information between multi-granularity regions for fine-grained feature learning and to directly model long-distance relationships for global understanding. The proposed method has been named the "Multi-Granularity and Multi-Attention Net (2M-Net)". Specifically, we first applied the jigsaw method to generate related tasks emphasizing regions with different granularities, and we then trained the model on these tasks using a hierarchical sharing mechanism. In effect, the training signals from the extra tasks created as an inductive bias, enabling 2M-Net to discover task relatedness without the need for annotations. Next, the self-attention mechanism acted as a plug-and-play module to effectively enhance the feature representation capabilities. Finally, multi-scale features were applied for prediction.

Results: A public data set of 14,236 hand radiographs, provided by the Radiological Society of North America (RSNA), was used to develop and validate 2M-Net. In the public benchmark testing, the mean absolute error (MAE) between the bone age estimates of the model and of the reviewer was 3.98 months (3.89 months for males and 4.07 months for females).

Conclusions: By using the jigsaw method to construct a multi-task learning strategy and inserting the self-attention module for efficient global modeling, we established 2M-Net, which is comparable to the previous best method in terms of performance.

Background: Amyotrophic lateral sclerosis (ALS)-related white-matter microstructural abnormalities have received considerable attention; however, gray-matter structural abnormalities have not been fully elucidated. This study aimed to evaluate cortical microstructural abnormalities in ALS and determine their association with disease severity.

Methods: This study included 34 patients with ALS and 30 healthy controls. Diffusion-weighted data were used to estimate neurite orientation dispersion and density imaging (NODDI) parameters, including neurite density index (NDI) and orientation dispersion index (ODI). We performed gray matter-based spatial statistics (GBSS) in a voxel-wise manner to determine the cortical microstructure difference. We used the revised ALS Functional Rating Scale (ALSFRS-R) to assess disease severity and conducted a correlation analysis between NODDI parameters and ALSFRS-R.

Results: In patients with ALS, the NDI reduction involved several cortical regions [primarily the precentral gyrus, postcentral gyrus, temporal cortex, prefrontal cortex, occipital cortex, and posterior parietal cortex; family-wise error (FWE)-corrected P<0.05]. ODI decreased in relatively few cortical regions (including the precentral gyrus, postcentral gyrus, prefrontal cortex, and inferior parietal lobule; FWE-corrected P<0.05). The NDI value in the left precentral and postcentral gyrus was positively correlated with the ALS disease severity (FWE-corrected P<0.05).

Conclusions: The decreases in NDI and ODI involved both motor-related and extra-motor regions and indicated the presence of gray-matter microstructural impairment in ALS. NODDI parameters are potential imaging biomarkers for evaluating disease severity in vivo. Our results showed that GBSS is a feasible method for identifying abnormalities in the cortical microstructure of patients with ALS.