Quantitative Imaging in Medicine and Surgery最新文献

Background: The incidence rate of thyroid nodules has reached 65%, but only 5-15% of these modules are malignant. Therefore, accurately determining the benign and malignant nature of thyroid nodules can prevent unnecessary treatment. We aimed to develop a deep-learning (DL) radiomics model based on ultrasound (US), explore its diagnostic efficacy for benign and malignant thyroid nodules, and verify whether it improved the diagnostic level of physicians.

Methods: We retrospectively included 1,076 thyroid nodules from 817 patients at three institutions. The radiomics and DL features of the US images were extracted and used to construct radiomics signature (Rad_sig) and deep-learning signature (DL_sig). A Pearson correlation analysis and least absolute shrinkage and selection operator (LASSO) regression analysis were used for feature selection. Clinical US semantic signature (C_US_sig) was constructed based on clinical information and US semantic features. Next, a combined model was constructed based on the above three signatures in the form of a nomogram. The model was constructed using a development set (institution 1: 719 nodules), and the model was evaluated using two external validation sets (institution 2: 74 nodules, and institution 3: 283 nodules). The performance of the model was assessed using decision curve analysis (DCA) and calibration curves. Furthermore, the C_US_sigs of junior physicians, senior physicians, and expers were constructed. The DL radiomics model was used to assist the physicians with different levels of experience in the interpretation of thyroid nodules.

Results: In the development and validation sets, the combined model showed the highest performance, with areas under the curve (AUCs) of 0.947, 0.917, and 0.929, respectively. The DCA results showed that the comprehensive nomogram had the best clinical utility. The calibration curves indicated good calibration for all models. The AUCs for distinguishing between benign and malignant thyroid nodules by junior physicians, senior physicians, and experts were 0.714-0.752, 0.740-0.824, and 0.891-0.908, respectively; however, with the assistance of DL radiomics, the AUCs reached 0.858-0.923, 0.888-0.944, and 0.912-0.919, respectively.

Conclusions: The nomogram based on DL radiomics had high diagnostic efficacy for thyroid nodules, and DL radiomics could assist physicians with different levels of experience to improve their diagnostic level.

Background: Axillary lymph node (ALN) status is a crucial prognostic indicator for breast cancer metastasis, with manual interpretation of whole slide images (WSIs) being the current standard practice. However, this method is subjective and time-consuming. Recent advancements in deep learning-based methods for medical image analysis have shown promise in improving clinical diagnosis. This study aims to leverage these technological advancements to develop a deep learning model based on features extracted from primary tumor biopsies for preoperatively identifying ALN metastasis in early-stage breast cancer patients with negative nodes.

Methods: We present DLCNBC-SA, a deep learning-based network specifically tailored for core needle biopsy and clinical data feature extraction, which integrates a self-attention mechanism (CNBC-SA). The proposed model consists of a feature extractor based on convolutional neural network (CNN) and an improved self-attention mechanism module, which can preserve the independence of features in WSIs for analysis and enhancement to provide rich feature representation. To validate the performance of the proposed model, we conducted comparative experiments and ablation studies using publicly available datasets, and verification was performed through quantitative analysis.

Results: The comparative experiment illustrates the superior performance of the proposed model in the task of binary classification of ALNs, as compared to alternative methods. Our method achieved outstanding performance [area under the curve (AUC): 0.882] in this task, significantly surpassing the state-of-the-art (SOTA) method on the same dataset (AUC: 0.862). The ablation experiment reveals that incorporating RandomRotation data augmentation technology and utilizing Adadelta optimizer can effectively enhance the performance of the proposed model.

Conclusions: The experimental results demonstrate that the model proposed in this paper outperforms the SOTA model on the same dataset, thereby establishing its reliability as an assistant for pathologists in analyzing WSIs of breast cancer. Consequently, it significantly enhances both the efficiency and accuracy of doctors during the diagnostic process.

Background: Persistent challenges associated with misdiagnosis and underdiagnosis of coronary microvascular disease (CMVD) necessitate the exploration of noninvasive imaging techniques to enhance diagnostic accuracy. Therefore, we aimed to integrate multimodal imaging approaches to achieve a higher diagnostic rate for CMVD using high-quality myocardial metabolism imaging (MMI) and myocardial contrast echocardiography (MCE). This combination diagnostic strategy may help address the urgent need for improved CMVD diagnosis.

Methods: In this study, we established five distinct pretreatment groups, each consisting of nine male rabbit: a fasted group, a nonfasted group, a sugar load group, an acipimox group, and a combination group of nonfasted rabbits administered insulin. Moreover, positron emission tomography-computed tomography (PET/CT) scan windows were established at 30-, 60-, and 90-minute intervals. We developed 10 CMVD models and conducted a diagnosis of CMVD through an integrated analysis of MMI and MCE, including image acquisition and processing. For each heart segment, we calculated the standardized uptake value (SUV) based on body weight (SUV_bw), as well as certain ratios of SUV including SUV of the heart (SUV_heart) to that of the liver (SUV_liver) and SUV_heart to SUV of the lung (SUV_lung). Additionally, we obtained three coronary SUV_bw uptake values. To clarify the relationship between SUV_bw uptake values and echocardiographic parameters of the myocardial contrast agent more thoroughly, we conducted a comprehensive analysis across different pretreatment protocols. Receiver operating characteristic (ROC) curve analysis was employed to evaluate the diagnostic accuracy of each parameter in the context of CMVD.

Results: In the context of MMI, the nonfasted-plus-insulin group, as observed during the 60-minute examination, exhibited a noteworthy total ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) uptake of 47.44±6.53 g/mL, which was found to be statistically different from the other groups. To ascertain the reliability of the results, two double-blind investigators independently assessed the data and achieved a good level of agreement, according to the intraclass correlation coefficient (ICC) (0.957). The SUV_bw of the nonfasted-plus-insulin group exhibited a moderate correlation with the microvascular blood flow reserve (MBFR) parameters derived from the MCE examination, as evidenced by a r value of 0.686. For the diagnosis of CMVD disease, the diagnostic accuracy of the combined diagnostic method [area under the curve (AUC) =0.789; 95% confidence interval (CI): 0.705-0.873] was significantly higher than that of the MBFR (AUC =0.697; 95% CI: 0.597-0.797) and SUV_bw (AUC =0.715; 95% CI: 0.622-0.807) methods (P<0.05).

Conclusions: Our study demonst

Background: Preoperative grading gliomas is essential for therapeutic clinical decision-making. Current non-invasive imaging modality for glioma grading were primarily focused on magnetic resonance imaging (MRI) or positron emission tomography (PET) of the tumor region. However, these methods overlook the peritumoral region (PTR) of tumor and cannot take full advantage of the biological information derived from hybrid-imaging. Therefore, we aimed to combine multiparameter from hybrid ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) PET/MRI of the solid component and PTR were combined for differentiating high-grade glioma (HGG) from low-grade glioma (LGG).

Methods: A total of 76 patients with pathologically confirmed glioma (41 HGG and 35 LGG) who underwent simultaneous ¹⁸F-FDG PET, arterial spin labelling (ASL), and diffusion-weighted imaging (DWI) with hybrid PET/MRI were retrospectively enrolled. The relative maximum standardized uptake value (rSUV_max), relative cerebral blood flow (rCBF), and relative minimum apparent diffusion coefficient (rADC_min) for the solid component and PTR at different distances outside tumoral border were compared. Receiver operating characteristic (ROC) curves were applied to assess the grading performance. A nomogram for HGG prediction was constructed.

Results: HGGs displayed higher rSUV_max and rCBF but lower rADC_min in the solid component and 5 mm-adjacent PTR, lower rADC_min in 10 mm-adjacent PTR, and higher rCBF in 15- and 20-mm-adjacent PTR. rSUV_max in solid component performed best [area under the curve (AUC) =0.865] as a single parameter for grading. Combination of rSUV_max in the solid component and adjacent 20 mm performed better (AUC =0.881). Integration of all 3 indicators in the solid component and adjacent 20 mm performed the best (AUC =0.928). The nomogram including rSUV_max, rCBF, and rADC_min in the solid component and 5-mm-adjacent PTR predicted HGG with a concordance index (C-index) of 0.906.

Conclusions: Multiparametric ¹⁸F-FDG PET/MRI from the solid component and PTR performed excellently in differentiating HGGs from LGGs. It can be used as a non-invasive and effective tool for preoperative grade stratification of patients with glioma, and can be considered in clinical practice.

Background: Extravaginal testicular torsion has profound clinical implications in neonates, but its ultrasound characteristics may vary at different disease stages. The purpose of this study was to identify the ultrasound characteristics of neonatal extravaginal testicular torsion and their diagnostic value at different disease stages.

Methods: A retrospective analysis of the clinical and ultrasound examination data of 20 infants aged 1-75 days with surgically and pathologically confirmed unilateral extravaginal testicular torsion (10 right, 10 left) was conducted. The infants were divided into three stages based on the ultrasound characteristics: double-ring effusion, calcification of the tunica vaginalis, and testicular atrophy.

Results: In the double-ring effusion stage, the affected testicles were enlarged with axial abnormalities, with the parenchymal testicular blood flow signal significantly reduced or absent. Twisted paratesticular masses and a "double-ring effusion sign" were visible. In the tunica vaginalis calcification stage, the affected testicles were slightly smaller, with axial abnormalities, absent blood flow signals in the testicular parenchyma, and strong echogenicity of the tunica vaginalis. In the testicular atrophy stage, the affected testicles were markedly smaller, with enhanced echogenicity in the tunica vaginalis and parenchyma, and absent blood flow signal in the testicular parenchyma. The volumes of the affected testicles gradually decreased from the stage of double-ring effusion to that of tunica vaginalis calcification, and then to testicular atrophy (P<0.05).

Conclusions: Neonatal extravaginal testicular torsion at different disease stages has distinct ultrasound features, and color doppler ultrasound plays an important role in the diagnosis and treatment of extravaginal testicular torsion.

Background: Low-dose computed tomography (LDCT) is a diagnostic imaging technique designed to minimize radiation exposure to the patient. However, this reduction in radiation may compromise computed tomography (CT) image quality, adversely impacting clinical diagnoses. Various advanced LDCT methods have emerged to mitigate this challenge, relying on well-matched LDCT and normal-dose CT (NDCT) image pairs for training. Nevertheless, these methods often face difficulties in distinguishing image details from nonuniformly distributed noise, limiting their denoising efficacy. Additionally, acquiring suitably paired datasets in the medical domain poses challenges, further constraining their applicability. Hence, the objective of this study was to develop an innovative denoising framework for LDCT images employing unpaired data.

Methods: In this paper, we propose a LDCT denoising network (DNCNN) that alleviates the need for aligning LDCT and NDCT images. Our approach employs generative adversarial networks (GANs) to learn and model the noise present in LDCT images, establishing a mapping from the pseudo-LDCT to the actual NDCT domain without the need for paired CT images.

Results: Within the domain of weakly supervised methods, our proposed model exhibited superior objective metrics on the simulated dataset when compared to CycleGAN and selective kernel-based cycle-consistent GAN (SKFCycleGAN): the peak signal-to-noise ratio (PSNR) was 43.9441, the structural similarity index measure (SSIM) was 0.9660, and the visual information fidelity (VIF) was 0.7707. In the clinical dataset, we conducted a visual effect analysis by observing various tissues through different observation windows. Our proposed method achieved a no-reference structural sharpness (NRSS) value of 0.6171, which was closest to that of the NDCT images (NRSS =0.6049), demonstrating its superiority over other denoising techniques in preserving details, maintaining structural integrity, and enhancing edge contrast.

Conclusions: Through extensive experiments on both simulated and clinical datasets, we demonstrated the superior efficacy of our proposed method in terms of denoising quality and quantity. Our method exhibits superiority over both supervised techniques, including block-matching and 3D filtering (BM3D), residual encoder-decoder convolutional neural network (RED-CNN), and Wasserstein generative adversarial network-VGG (WGAN-VGG), and over weakly supervised approaches, including CycleGAN and SKFCycleGAN.

Background: Noninvasive evaluation of fetal lung development is a critical area of study. Two-dimensional shear-wave elastography (2D-SWE) provides valuable insights into tissue stiffness, potentially correlating with different stages of lung development. This study aims to explore the potential of the 2D-SWE technique for assessing the maturity of fetal lung development.

Methods: This prospective cohort study included pregnant women undergoing routine antenatal ultrasound examinations at the Second Affiliated Hospital of Fujian Medical University and Quanzhou Women's and Children's Hospital from September 2022 to September 2023. The study consecutively recruited 300 pregnant women with normal pregnancies and 15 who opted for induced labor. Among those with normal pregnancies, the study assessed the differences in fetal pulmonary and hepatic elasticity measurements across different gestational weeks (GW) using one-way analysis of variance (ANOVA). Furthermore, regression analyses using linear, quadratic, and cubic equations were conducted to investigate the relationship between fetal parameters and GW. For those who opted for induced labor, elasticity measurements were taken before induction, and fetal lung tissue specimens were collected for post-induction observation.

Results: Fetal lung and liver elasticity values, along with the lung-to-liver elasticity ratio (LLE ratio), showed significant variations across different GW (P<0.05). Specifically, fetal lung elasticity values initially increased and then decreased as GW advanced (R2=0.41). Liver elasticity values continuously increased throughout GW, though the rate of increase diminished during the prenatal period (R2=0.37). The LLE ratio values increased and then decreased over GW, fluctuating overall between 0.8 and 0.9 (R2=0.14). A 71.4% concordance was observed between the predicted stage of lung development, based on lung elasticity values, and the histological stage of lung development in the induced fetuses.

Conclusions: 2D-SWE can depict the maturation of fetal lung development at various stages.

Background: Barlow's disease (BD) is a common etiology of degenerative mitral valve (MV) disease, often causing significant mitral regurgitation (MR). The pathology of BD is challenging for surgeons performing MV repair (MVR). However, most MVR effectiveness studies have been based on survival and risk of reoperation. The aim of this study was to analyze the safety, efficacy, and durability of MVR in patients with BD and to identify factors that influence recurrent MR.

Methods: We retrospectively analyzed the clinical outcomes of 274 patients undergoing MVR for BD at a tertiary hospital (Guangdong People's Hospital, Guangzhou, China) between January 2010 and June 2022. To analyze the results of MVR and identify the risk factors for MR recurrence, we defined two groups: a total of 240 patients with MR grade <2+ (group A) and a total of 34 patients who had recurrent MR after MVR (group B; the patients with MR ≥2+). All patients were operated on using standard repair techniques. Recurrent MR was the primary outcome. Secondary outcomes were death and reoperation after MVR. Patients were followed up until March 2023. Patients were followed up by clinic visits, telephone calls, and postal or electronic questionnaires.

Results: The median [range] patient age was 46.00 [16-75] years and 186 (67.9%) patients were male. Concomitant procedures were performed in 123 patients: tricuspid valve repair 71 (25.9%), maze or pulmonary vein isolation (PVI) 12 (4.4%), atrial septal defect (ASD) repair 3 (1.1%), and left atrial appendage (LAA) closure 28 (10.2%). Hospital mortality was 0.4%. Long-term complications included radiofrequency ablation in 7 patients (2.6%), pacemaker implantation in 1 patient (0.4%), and stroke in 3 patients (1.1%). The median follow-up was 3.28 (range, 0-12.39) years. Considering the competing risk of mortality, the cumulative incidence of MR progression 2+ or more grades was 2.6%, 5.9%, 14.5%, and 27.7% at 1 month, 1, 5, and 10 years, respectively. Overall survival at 1, 5, and 10 years was 99.3%, 98.6%, and 98.6%, respectively. The immediate postoperative MR area [hazard ratio (HR) =1.723; 95% confidence interval (CI): 1.051-2.824; P=0.031], postoperative left ventricular end-diastolic dimension (LVEDD) (HR =1.149; 95% CI: 1.016-1.300; P=0.027), and postoperative MR grade {HR = Exp[4.500 - 0.544 × ln(t + 20)]; P=0.008} were associated with an increased risk of MR recurrence, whereas a higher left ventricular ejection fraction (LVEF) (HR =0.931; 95% CI: 0.868-0.999; P=0.049) was associated with a decreased risk.

Conclusions: MVR in patients with BD can be performed with low mortality and complications and is associated with superior long-term outcomes. However, MVR was associated with a certain risk of MR recurrence, especially in those with high postoperative LVEDD, residual MR >1+, and decreased postoperative LVEF. We recommend MVR for patients wit

Background: The contrasted-enhanced ultrasound thyroid imaging reporting and data system (CEUS TI-RADS) is the first international risk stratification system for thyroid nodules based on conventional ultrasound (US) and CEUS. This study aimed to evaluate the diagnostic efficacy of CEUS TI-RADS for benign and malignant thyroid nodules and to assess the related interobserver agreement.

Methods: The study recruited 433 patients who underwent thyroid US and CEUS between January 2019 and June 2023 at the Affiliated Hospital of Guangdong Medical University. A retrospective analysis of 467 thyroid nodules confirmed by fine-needle aspiration (FNA) and/or surgery was performed. Further, a CEUS TI-RADS classification was assigned to each thyroid nodule based on the CEUS TI-RADS scoring criteria for the US and CEUS features of the nodule. The nodules were grouped based on their sizes as follows: size ≤1 cm, group A; size >1 and ≤4 cm, group B; and size >4 cm, group C. Multivariate logistic regression was used to analyze independent risk factors for malignant thyroid nodules. Pathological assessment was the reference standard for establishing the sensitivity (SEN), specificity (SPE), accuracy (ACC), positive predictive value (PPV), and negative predictive value (NPV) of CEUS TI-RADS in diagnosing malignant thyroid nodules. The area under the curve (AUC) in the receiver operating characteristic (ROC) curve analysis was used to compare the diagnostic efficacy of the scoring system in predicting malignancy in three groups of nodules. The intragroup correlation coefficient (ICC) was adopted to assess the interobserver agreement of the CEUS TI-RADS score.

Results: Out of the 467 thyroid nodules, 262 were malignant and 205 were benign. Logistic regression analysis revealed that the independent risk factors for malignant thyroid nodules included punctate echogenic foci (P<0.001), taller-than-wide shape (P=0.015), extrathyroidal invasion (P=0.020), irregular margins/lobulation (P=0.036), hypoechoicity on US (P=0.038), and hypoenhancement on CEUS (P<0.001). The AUC for the CEUS TI-RADS in diagnosing malignant thyroid nodules was 0.898 for all nodules, 0.795 for group A, 0.949 for group B, and 0.801 for group C, with the optimal cutoff values of the CEUS TI-RADS being 5 points, 6 points, 5 points, and 5 points, respectively. Among these groups of nodules, group B had the highest AUC, with the SEN, SPE, ACC, PPV, and NPV for diagnosing malignant nodules being 95.9%, 88.1%, 92.8%, 92.6%, and 93.2%, respectively. The ICC of the CEUS TI-RADS classification between senior and junior physicians was 0.862 (P<0.001).

Conclusions: In summary, CEUS TI-RADS demonstrated significant efficacy in distinguishing thyroid nodules. Nonetheless, there were variations in its capacity to detect malignant nodules across diverse sizes, and it demonstrate optimal performance in 1- to 4-cm nodules. These

Background: Previous studies have indicated that despite adhering to current patient selection guidelines, there remains a 30% to 40% subset of patients who do not experience improvement in heart failure (HF) after receiving cardiac resynchronization therapy (CRT). We aim to utilize echocardiographic myocardial work parameters to serve as predictors of responsiveness to CRT in patients with heart failure and reduced ejection fraction (HFrEF).

Methods: We prospectively recruited patients who underwent CRT at Sun Yat-sen Memorial Hospital from June 2019 to September 2022. Comprehensive preoperative information, clinical laboratory data, conventional echocardiographic parameters and myocardial work were collected for all participants, as well as follow-up data 6 months after CRT.

Results: Twenty-five patients (67.6%) showed response to CRT treatment, while twelve patients (32.4%) had no response. Compared with the non-response group, the response group had larger region constructive work [RCW: the sum of constructive work (CW) in the 9 segments of the basal, mid, and apical segments of the anterior, lateral, and posterior walls], region wasted work [RWW: the sum of wasted work (WW) in the 6 segments of the basal and mid segments of the anterior septum, posterior septum and anterior walls], and the combination of RCW and RWW (RCW + RWW) in baseline (RCW: 9,695.68±2,955.40 vs. 5,219.50±2,207.68 mmHg%, P<0.001; RWW: 3,612.08±1,723.80 vs. 1,674.33±995.23 mmHg%, P=0.001; RCW + RWW: 13,307.76±3,857.71 vs. 6,893.83±2,592.83 mmHg%, P<0.001). Furthermore, global constructive work (GCW), global wasted work (GWW), GCW + GWW, RCW, RWW, and RCW + RWW had areas under the receiver operating characteristic curve (AUCs) of 0.870, 0.770, 0.860, 0.890, 0.870, and 0.910, respectively, for predicting CRT responsiveness.

Conclusions: The global and regional myocardial work parameters are associated with CRT response in CRT candidates. Particularly regional myocardial work parameters appear to be promising parameters to improve selection for CRT of patients with HFrEF.