Cascaded cross-attention transformers and convolutional neural networks for multi-organ segmentation in male pelvic computed tomography.

Purpose: Segmentation of the prostate and surrounding organs at risk from computed tomography is required for radiation therapy treatment planning. We propose an automatic two-step deep learning-based segmentation pipeline that consists of an initial multi-organ segmentation network for organ localization followed by organ-specific fine segmentation. Approach: Initial segmentation of all target organs is performed using a hybrid convolutional-transformer model, axial cross-attention UNet. The output from this model allows for region of interest computation and is used to crop tightly around individual organs for organ-specific fine segmentation. Information from this network is also propagated to the fine segmentation stage through an image enhancement module, highlighting regions of interest in the original image that might be difficult to segment. Organ-specific fine segmentation is performed on these cropped and enhanced images to produce the final output segmentation. Results: We apply the proposed approach to segment the prostate, bladder, rectum, seminal vesicles, and femoral heads from male pelvic computed tomography (CT). When tested on a held-out test set of 30 images, our two-step pipeline outperformed other deep learning-based multi-organ segmentation algorithms, achieving average dice similarity coefficient (DSC) of 0.836±0.071 (prostate), 0.947±0.038 (bladder), 0.828±0.057 (rectum), 0.724±0.101 (seminal vesicles), and 0.933±0.020 (femoral heads). Conclusions: Our results demonstrate that a two-step segmentation pipeline with initial multi-organ segmentation and additional fine segmentation can delineate male pelvic CT organs well. The utility of this additional layer of fine segmentation is most noticeable in challenging cases, as our two-step pipeline produces noticeably more accurate and less erroneous results compared to other state-of-the-art methods on such images.

Web-Based Application for Biomedical Image Registry, Analysis, and Translation (BiRAT).

Imaging has become an invaluable tool in preclinical research for its capability to non-invasively detect and monitor disease and assess treatment response. With the increased use of preclinical imaging, large volumes of image data are being generated requiring critical data management tools. Due to proprietary issues and continuous technology development, preclinical images, unlike DICOM-based images, are often stored in an unstructured data file in company-specific proprietary formats. This limits the available DICOM-based image management database to be effectively used for preclinical applications. A centralized image registry and management tool is essential for advances in preclinical imaging research. Specifically, such tools may have a high impact in generating large image datasets for the evolving artificial intelligence applications and performing retrospective analyses of previously acquired images. In this study, a web-based server application is developed to address some of these issues. The application is designed to reflect the actual experimentation workflow maintaining detailed records of both individual images and experimental data relevant to specific studies and/or projects. The application also includes a web-based 3D/4D image viewer to easily and quickly view and evaluate images. This paper briefly describes the initial implementation of the web-based application.

Shortwave Infrared-Emitting Theranostics for Breast Cancer Therapy Response Monitoring.

Therapeutic drug monitoring (TDM) in cancer, while imperative, has been challenging due to inter-patient variability in drug pharmacokinetics. Additionally, most pharmacokinetic monitoring is done by assessments of the drugs in plasma, which is not an accurate gauge for drug concentrations in target tumor tissue. There exists a critical need for therapy monitoring tools that can provide real-time feedback on drug efficacy at target site to enable alteration in treatment regimens early during cancer therapy. Here, we report on theranostic optical imaging probes based on shortwave infrared (SWIR)-emitting rare earth-doped nanoparticles encapsulated with human serum albumin (abbreviated as ReANCs) that have demonstrated superior surveillance capability for detecting micro-lesions at depths of 1 cm in a mouse model of breast cancer metastasis. Most notably, ReANCs previously deployed for detection of multi-organ metastases resolved bone lesions earlier than contrast-enhanced magnetic resonance imaging (MRI). We engineered tumor-targeted ReANCs carrying a therapeutic payload as a potential theranostic for evaluating drug efficacy at the tumor site. In vitro results demonstrated efficacy of ReANCs carrying doxorubicin (Dox), providing sustained release of Dox while maintaining cytotoxic effects comparable to free Dox. Significantly, in a murine model of breast cancer lung metastasis, we demonstrated the ability for therapy monitoring based on measurements of SWIR fluorescence from tumor-targeted ReANCs. These findings correlated with a reduction in lung metastatic burden as quantified via MRI-based volumetric analysis over the course of four weeks. Future studies will address the potential of this novel class of theranostics as a preclinical pharmacological screening tool.