Semi-supervised semantic segmentation of prostate and organs-at-risk on 3D pelvic CT images

Generative AI for synthetic data across multiple medical modalities: A systematic review of recent developments and challenges

This paper presents a comprehensive systematic review of generative models (GANs, VAEs, DMs, and LLMs) used to synthesize various medical data types, including imaging (dermoscopic, mammographic, ultrasound, CT, MRI, and X-ray), text, time-series, and tabular data (EHR). Unlike previous narrowly focused reviews, our study encompasses a broad array of medical data modalities and explores various generative models. Our aim is to offer insights into their current and future applications in medical research, particularly in the context of synthesis applications, generation techniques, and evaluation methods, as well as providing a GitHub repository as a dynamic resource for ongoing collaboration and innovation. Our search strategy queries databases such as Scopus, PubMed, and ArXiv, focusing on recent works from January 2021 to November 2023, excluding reviews and perspectives. This period emphasizes recent advancements beyond GANs, which have been extensively covered in previous reviews. The survey also emphasizes the aspect of conditional generation, which is not focused on in similar work. Key contributions include a broad, multi-modality scope that identifies cross-modality insights and opportunities unavailable in single-modality surveys. While core generative techniques are transferable, we find that synthesis methods often lack sufficient integration of patient-specific context, clinical knowledge, and modality-specific requirements tailored to the unique characteristics of medical data. Conditional models leveraging textual conditioning and multimodal synthesis remain underexplored but offer promising directions for innovation. Our findings are structured around three themes: (1) Synthesis applications, highlighting clinically valid synthesis applications and significant gaps in using synthetic data beyond augmentation, such as for validation and evaluation; (2) Generation techniques, identifying gaps in personalization and cross-modality innovation; and (3) Evaluation methods, revealing the absence of standardized benchmarks, the need for large-scale validation, and the importance of privacy-aware, clinically relevant evaluation frameworks. These findings emphasize the need for benchmarking and comparative studies to promote openness and collaboration.

Deep learning based clinical target volumes contouring for prostate cancer: Easy and efficient application

BACKGROUND

Radiotherapy has been crucial in prostate cancer treatment. However, manual segmentation is labor intensive and highly variable among radiation oncologists. In this study, a deep learning based automated contouring model is constructed for clinical target volumes (CTVs) of intact and postoperative prostate cancer.

METHODS

Computed tomography (CT) data sets of 197 prostate cancer patients were collected. Two auto-delineation models were built for radical radiotherapy and postoperative radiotherapy of prostate cancer respectively, and each model included CTVn for pelvic lymph nodes and CTVp for prostate tumors or prostate tumor beds.

RESULTS

In the radical radiotherapy model, the volumetric dice (VD) coefficient of CTVn calculated by AI, was higher than that of the one delineated by the junior physicians (0.85 vs. 0.82, p = 0.018); In the postoperative radiotherapy model, the quantitative parameter of CTVn and CTVp, counted by AI, was better than that of the junior physicians. The median delineation time for AI was 0.23 min in the postoperative model and 0.26 min in the radical model, which were significantly shorter than those of the physicians (50.40 and 45.43 min, respectively, p < 0.001). The correction time of the senior physician for AI was much shorter compared with that for the junior physicians in both models (p < 0.001).

CONCLUSION

Using deep learning and attention mechanism, a highly consistent and time-saving contouring model was built for CTVs of pelvic lymph nodes and prostate tumors or prostate tumor beds for prostate cancer, which also might be a good approach to train junior radiation oncologists.

GAN-based generation of realistic 3D volumetric data: A systematic review and taxonomy

With the massive proliferation of data-driven algorithms, such as deep learning-based approaches, the availability of high-quality data is of great interest. Volumetric data is very important in medicine, as it ranges from disease diagnoses to therapy monitoring. When the dataset is sufficient, models can be trained to help doctors with these tasks. Unfortunately, there are scenarios where large amounts of data is unavailable. For example, rare diseases and privacy issues can lead to restricted data availability. In non-medical fields, the high cost of obtaining enough high-quality data can also be a concern. A solution to these problems can be the generation of realistic synthetic data using Generative Adversarial Networks (GANs). The existence of these mechanisms is a good asset, especially in healthcare, as the data must be of good quality, realistic, and without privacy issues. Therefore, most of the publications on volumetric GANs are within the medical domain. In this review, we provide a summary of works that generate realistic volumetric synthetic data using GANs. We therefore outline GAN-based methods in these areas with common architectures, loss functions and evaluation metrics, including their advantages and disadvantages. We present a novel taxonomy, evaluations, challenges, and research opportunities to provide a holistic overview of the current state of volumetric GANs.

Towards Automation in Radiotherapy Planning: A Deep Learning Approach for the Delineation of Parotid Glands in Head and Neck Cancer

The delineation of parotid glands in head and neck (HN) carcinoma is critical to assess radiotherapy (RT) planning. Segmentation processes ensure precise target position and treatment precision, facilitate monitoring of anatomical changes, enable plan adaptation, and enhance overall patient safety. In this context, artificial intelligence (AI) and deep learning (DL) have proven exceedingly effective in precisely outlining tumor tissues and, by extension, the organs at risk. This paper introduces a DL framework using the AttentionUNet neural network for automatic parotid gland segmentation in HN cancer. Extensive evaluation of the model is performed in two public and one private dataset, while segmentation accuracy is compared with other state-of-the-art DL segmentation schemas. To assess replanning necessity during treatment, an additional registration method is implemented on the segmentation output, aligning images of different modalities (Computed Tomography (CT) and Cone Beam CT (CBCT)). AttentionUNet outperforms similar DL methods (Dice Similarity Coefficient: 82.65% ± 1.03, Hausdorff Distance: 6.24 mm ± 2.47), confirming its effectiveness. Moreover, the subsequent registration procedure displays increased similarity, providing insights into the effects of RT procedures for treatment planning adaptations. The implementation of the proposed methods indicates the effectiveness of DL not only for automatic delineation of the anatomical structures, but also for the provision of information for adaptive RT support.

Semi-Supervised Medical Image Segmentation Guided by Bi-Directional Constrained Dual-Task Consistency

BACKGROUND

Medical image processing tasks represented by multi-object segmentation are of great significance for surgical planning, robot-assisted surgery, and surgical safety. However, the exceptionally low contrast among tissues and limited available annotated data makes developing an automatic segmentation algorithm for pelvic CT challenging.

METHODS

A bi-direction constrained dual-task consistency model named PICT is proposed to improve segmentation quality by leveraging free unlabeled data. First, to learn more unmarked data features, it encourages the model prediction of the interpolated image to be consistent with the interpolation of the model prediction at the pixel, model, and data levels. Moreover, to constrain the error prediction of interpolation interference, PICT designs an auxiliary pseudo-supervision task that focuses on the underlying information of non-interpolation data. Finally, an effective loss algorithm for both consistency tasks is designed to ensure the complementary manner and produce more reliable predictions.

RESULTS

Quantitative experiments show that the proposed PICT achieves 87.18%, 96.42%, and 79.41% mean DSC score on ACDC, CTPelvic1k, and the individual Multi-tissue Pelvis dataset with gains of around 0.8%, 0.5%, and 1% compared to the state-of-the-art semi-supervised method. Compared to the baseline supervised method, the PICT brings over 3-9% improvements.

CONCLUSIONS

The developed PICT model can effectively leverage unlabeled data to improve segmentation quality of low contrast medical images. The segmentation result could improve the precision of surgical path planning and provide input for robot-assisted surgery.

Semisupervised Semantic Segmentation with Mutual Correction Learning

The semisupervised semantic segmentation method uses unlabeled data to effectively reduce the required labeled data, and the pseudo supervision performance is greatly influenced by pseudo labels. Therefore, we propose a semisupervised semantic segmentation method based on mutual correction learning, which effectively corrects the wrong convergence direction of pseudo supervision. The well-calibrated segmentation confidence maps are generated through the multiscale feature fusion attention mechanism module. More importantly, using internal knowledge, a mutual correction mechanism based on consistency regularization is proposed to correct the convergence direction of pseudo labels during cross pseudo supervision. The multiscale feature fusion attention mechanism module and mutual correction learning improve the accuracy of the entire learning process. Experiments show that the MIoU (mean intersection over union) reaches 75.32%, 77.80%, 78.95%, and 79.16% using 1/16, 1/8, 1/4, and 1/2 labeled data on PASCAL VOC 2012. The results show that the new approach achieves an advanced level.

Weaving attention U-net: A novel hybrid CNN and attention-based method for organs-at-risk segmentation in head and neck CT images

PURPOSE

In radiotherapy planning, manual contouring is labor-intensive and time-consuming. Accurate and robust automated segmentation models improve the efficiency and treatment outcome. We aim to develop a novel hybrid deep learning approach, combining convolutional neural networks (CNNs) and the self-attention mechanism, for rapid and accurate multi-organ segmentation on head and neck computed tomography (CT) images.

METHODS

Head and neck CT images with manual contours of 115 patients were retrospectively collected and used. We set the training/validation/testing ratio to 81/9/25 and used the 10-fold cross-validation strategy to select the best model parameters. The proposed hybrid model segmented 10 organs-at-risk (OARs) altogether for each case. The performance of the model was evaluated by three metrics, that is, the Dice Similarity Coefficient (DSC), Hausdorff distance 95% (HD95), and mean surface distance (MSD). We also tested the performance of the model on the head and neck 2015 challenge dataset and compared it against several state-of-the-art automated segmentation algorithms.

RESULTS

The proposed method generated contours that closely resemble the ground truth for 10 OARs. On the head and neck 2015 challenge dataset, the DSC scores of these OARs were 0.91 ± 0.02, 0.73 ± 0.10, 0.95 ± 0.03, 0.76 ± 0.08, 0.79 ± 0.05, 0.87 ± 0.05, 0.86 ± 0.08, 0.87 ± 0.03, and 0.87 ± 0.07 for brain stem, chiasm, mandible, left/right optic nerve, left/right submandibular, and left/right parotid, respectively. Our results of the new weaving attention U-net (WAU-net) demonstrate superior or similar performance on the segmentation of head and neck CT images.

CONCLUSIONS

We developed a deep learning approach that integrates the merits of CNNs and the self-attention mechanism. The proposed WAU-net can efficiently capture local and global dependencies and achieves state-of-the-art performance on the head and neck multi-organ segmentation task.