Computerized Medical Imaging and Graphics最新文献

Despite that deep learning has achieved state-of-the-art performance for automatic medical image segmentation, it often requires a large amount of pixel-level manual annotations for training. Obtaining these high-quality annotations is time-consuming and requires specialized knowledge, which hinders the widespread application that relies on such annotations to train a model with good segmentation performance. Using scribble annotations can substantially reduce the annotation cost, but often leads to poor segmentation performance due to insufficient supervision. In this work, we propose a novel framework named as ScribSD+ that is based on multi-scale knowledge distillation and class-wise contrastive regularization for learning from scribble annotations. For a student network supervised by scribbles and the teacher based on Exponential Moving Average (EMA), we first introduce multi-scale prediction-level Knowledge Distillation (KD) that leverages soft predictions of the teacher network to supervise the student at multiple scales, and then propose class-wise contrastive regularization which encourages feature similarity within the same class and dissimilarity across different classes, thereby effectively improving the segmentation performance of the student network. Experimental results on the ACDC dataset for heart structure segmentation and a fetal MRI dataset for placenta and fetal brain segmentation demonstrate that our method significantly improves the student’s performance and outperforms five state-of-the-art scribble-supervised learning methods. Consequently, the method has a potential for reducing the annotation cost in developing deep learning models for clinical diagnosis.

Purpose

To evaluate lymphovascular invasion (LVI) in breast cancer by comparing the diagnostic performance of preoperative multimodal magnetic resonance imaging (MRI)-based radiomics and deep-learning (DL) models.

Methods

This retrospective study included 262 patients with breast cancer—183 in the training cohort (144 LVI-negative and 39 LVI-positive cases) and 79 in the validation cohort (59 LVI-negative and 20 LVI-positive cases). Radiomics features were extracted from the intra- and peritumoral breast regions using multimodal MRI to generate gross tumor volume (GTV)_radiomics and gross tumor volume plus peritumoral volume (GPTV)_radiomics. Subsequently, DL models (GTV_DL and GPTV_DL) were constructed based on the GTV and GPTV to determine the LVI status. Finally, the most effective radiomics and DL models were integrated with imaging findings to establish a hybrid model, which was converted into a nomogram to quantify the LVI risk.

Results

The diagnostic efficiency of GPTV_DL was superior to that of GTV_DL (areas under the curve [AUCs], 0.771 and 0.720, respectively). Similarly, GPTV_radiomics outperformed GTV_radiomics (AUC, 0.685 and 0.636, respectively). Univariate and multivariate logistic regression analyses revealed an association between imaging findings, such as MRI-axillary lymph nodes and peritumoral edema (AUC, 0.665). The hybrid model achieved the highest accuracy by combining GPTV_DL, GPTV_radiomics, and imaging findings (AUC, 0.872).

Conclusion

The diagnostic efficiency of the GPTV-derived radiomics and DL models surpassed that of the GTV-derived models. Furthermore, the hybrid model, which incorporated GPTV_DL, GPTV_radiomics, and imaging findings, demonstrated the effective determination of LVI status prior to surgery in patients with breast cancer.

The use of multi-modality non-contrast images (i.e., T1FS, T2FS and DWI) for segmenting liver tumors provides a solution by eliminating the use of contrast agents and is crucial for clinical diagnosis. However, this remains a challenging task to discover the most useful information to fuse multi-modality images for accurate segmentation due to inter-modal interference. In this paper, we propose a dual-stream multi-level fusion framework (DM-FF) to, for the first time, accurately segment liver tumors from non-contrast multi-modality images directly. Our DM-FF first designs an attention-based encoder–decoder to effectively extract multi-level feature maps corresponding to a specified representation of each modality. Then, DM-FF creates two types of fusion modules, in which a module fuses learned features to obtain a shared representation across multi-modality images to exploit commonalities and improve the performance, and a module fuses the decision evidence of segment to discover differences between modalities to prevent interference caused by modality’s conflict. By integrating these three components, DM-FF enables multi-modality non-contrast images to cooperate with each other and enables an accurate segmentation. Evaluation on 250 patients including different types of tumors from two MRI scanners, DM-FF achieves a Dice of 81.20%, and improves performance (Dice by at least 11%) when comparing the eight state-of-the-art segmentation architectures. The results indicate that our DM-FF significantly promotes the development and deployment of non-contrast liver tumor technology.

Despite sharing the same histologic classification, individual tumors in multi metastatic patients may present with different characteristics and varying sensitivities to anticancer therapies. In this study, we investigate the utility of radiomic biomarkers for prediction of lesion-specific treatment resistance in multi metastatic leiomyosarcoma patients. Using a dataset of n=202 lung metastases (LM) from n=80 patients with 1648 pre-treatment computed tomography (CT) radiomics features and LM progression determined from follow-up CT, we developed a radiomic model to predict the progression of each lesion. Repeat experiments assessed the relative predictive performance across LM volume groups. Lesion-specific radiomic models indicate up to a 4.5-fold increase in predictive capacity compared with a no-skill classifier, with an area under the precision-recall curve of 0.70 for the most precise model (FDR = 0.05). Precision varied by administered drug and LM volume. The effect of LM volume was controlled by removing radiomic features at a volume-correlation coefficient threshold of 0.20. Predicting lesion-specific responses using radiomic features represents a novel strategy by which to assess treatment response that acknowledges biological diversity within metastatic subclones, which could facilitate management strategies involving selective ablation of resistant clones in the setting of systemic therapy.

Pelvic fracture is a complex and severe injury. Accurate diagnosis and treatment planning require the segmentation of the pelvic structure and the fractured fragments from preoperative CT scans. However, this segmentation is a challenging task, as the fragments from a pelvic fracture typically exhibit considerable variability and irregularity in the morphologies, locations, and quantities. In this study, we propose a novel dual-stream learning framework for the automatic segmentation and category labeling of pelvic fractures. Our method uniquely identifies pelvic fracture fragments in various quantities and locations using a dual-branch architecture that leverages distance learning from bone fragments. Moreover, we develop a multi-size feature fusion module that adaptively aggregates features from diverse receptive fields tailored to targets of different sizes and shapes, thus boosting segmentation performance. Extensive experiments on three pelvic fracture datasets from different medical centers demonstrated the accuracy and generalizability of the proposed method. It achieves a mean Dice coefficient and mean Sensitivity of 0.935$\pm$0.068 and 0.929$\pm$0.058 in the dataset FracCLINIC, and 0.955$\pm$0.072 and 0.912$\pm$0.125 in the dataset FracSegData, which are superior than other comparing methods. Our method optimizes the process of pelvic fracture segmentation, potentially serving as an effective tool for preoperative planning in the clinical management of pelvic fractures.

Background

Radiation therapy is one of the crucial treatment modalities for cancer. An excellent radiation therapy plan relies heavily on an outstanding dose distribution map, which is traditionally generated through repeated trials and adjustments by experienced physicists. However, this process is both time-consuming and labor-intensive, and it comes with a degree of subjectivity. Now, with the powerful capabilities of deep learning, we are able to predict dose distribution maps more accurately, effectively overcoming these challenges.

Methods

In this study, we propose a novel Swin-UMamba-Channel prediction model specifically designed for predicting the dose distribution of patients with left breast cancer undergoing radiotherapy after total mastectomy. This model integrates anatomical position information of organs and ray angle information, significantly enhancing prediction accuracy. Through iterative training of the generator (Swin-UMamba) and discriminator, the model can generate images that closely match the actual dose, assisting physicists in quickly creating DVH curves and shortening the treatment planning cycle. Our model exhibits excellent performance in terms of prediction accuracy, computational efficiency, and practicality, and its effectiveness has been further verified through comparative experiments with similar networks.

Results

The results of the study indicate that our model can accurately predict the clinical dose of breast cancer patients undergoing intensity-modulated radiation therapy (IMRT). The predicted dose range is from 0 to 50 Gy, and compared with actual data, it shows a high accuracy with an average Dice similarity coefficient of 0.86. Specifically, the average dose change rate for the planning target volume ranges from 0.28 % to 1.515 %, while the average dose change rates for the right and left lungs are 2.113 % and 0.508 %, respectively. Notably, due to their small sizes, the heart and spinal cord exhibit relatively higher average dose change rates, reaching 3.208 % and 1.490 %, respectively. In comparison with similar dose studies, our model demonstrates superior performance. Additionally, our model possesses fewer parameters, lower computational complexity, and shorter processing time, further enhancing its practicality and efficiency. These findings provide strong evidence for the accuracy and reliability of our model in predicting doses, offering significant technical support for IMRT in breast cancer patients.

Conclusion

This study presents a novel Swin-UMamba-Channel dose prediction model, and its results demonstrate its precise prediction of clinical doses for the target area of left breast cancer patients undergoing total mastectomy and IMRT. These remarkable achievements provide valuable reference data for subsequent plan optimization and quality control, paving a new path for the application of deep learning in