Physical and Engineering Sciences in Medicine

Longitudinal deep learning models for tracking disease progression in ovarian cancer using PET/CT imaging and clinical reports

Ovarian cancer is often diagnosed at advanced stages, with high-grade serous ovarian cancer (HGSOC) accounting for 70-80% of fatalities. Current predictive tools, limited by single-time-point data, fail to capture subtle temporal changes indicative of relapse. To evaluate the performance of OvarXNet, a novel deep learning framework integrating longitudinal PET/CT imaging and clinical data for early prediction of ovarian cancer relapse. This retrospective study included 58 advanced-stage HGSOC patients (mean age, 56 ± 10.4 years) who underwent [

Potential anatomical triggers for plan adaptation of cervical cancer external beam radiotherapy

Abstract This study aimed to identify potential anatomical variation triggers using magnetic resonance imaging for plan adaption of cervical cancer patients to ensure dose requirements were met over an external beam radiotherapy course. Magnetic resonance images (MRIs) acquired before and during treatment were rigidly registered to a pre-treatment computerised tomography (CT) image for 11 retrospective cervix cancer datasets. Target volumes (TVs) and organs at risk (OARs) were delineated on both MRIs and propagated onto the CT. Treatment plans were generated based on the pre-treatment contours and applied to the mid-treatment contours. Anatomical and dosimetric changes between each timepoint were assessed. The anatomical changes included the change in centroid position and volume size. Dosimetric changes included the V30Gy and V40Gy for the OARs, and V95%, V100%, D95% and D98% for the TVs. Correlation with dosimetric and anatomical changes were assessed to determine potential replan triggers. Changes in the bowel volume and position in the superior-inferior direction, and the high-risk CTV anterior posterior position were highly correlated with a change in dose to the bowel and target, respectively. Hence changes in bowel and high-risk CTV could be used as a potential replan triggers.

Enhanced detection of ovarian cancer using AI-optimized 3D CNNs for PET/CT scan analysis

This study investigates how deep learning (DL) can enhance ovarian cancer diagnosis and staging using large imaging datasets. Specifically, we compare six conventional convolutional neural network (CNN) architectures-ResNet, DenseNet, GoogLeNet, U-Net, VGG, and AlexNet-with OCDA-Net, an enhanced model designed for [

Feasibility of the 3D printer to design an intracavitary applicator for the treatment of cervical cancer patients with high dose rate brachytherapy system

Designing an intracavitary brachytherapy applicator with a 3D printer using Polyamide12 for Tandem-Ovoid configuration. Further, to evaluate its feasibility and initial clinical use for the treatment of cervical cancer patients with a High Dose Rate (HDR) brachytherapy system. SolidWorks, Computer Aided design software was used for the design of the intracavitary brachytherapy applicator. Hewlett-Packard Jet Fusion 4200 was used for printing different parts of applicators with Polyamide12 (PA12) material. Radiograph and CT images of printed material parts were taken in the air and water medium to see the visualization. Before use in the patient, necessary quality assurance tests were carried out by coupling it with a microSelectron HDR machine. X-ray markers were used to visualize the source path inside the uterine and vaginal tandems. Physical and clinical evaluations were performed with a prototype 3D-printed applicator to check its suitability for clinical use. Final Applicator design was created from multiple hit and trial methods in SolidWorks. Printed PA12 of ovoid parts having a mean Hounsfield unit (HU) value of - 75 HU. Quality tests on the PA12 intracavitary applicator performed with the microSelectron HDR brachytherapy machine were passed. The chances of uterine perforation were less due to the semi-rigidity of the PA12 applicator. The newly designed T-O-based applicator and dummy marker do not produce any artifacts on the CT images. A low-cost flexible plastic applicator was developed that allowed the user to guide the tandem into the uterus of a patient. The developed PA12 intracavitary brachytherapy applicator did not produce artifacts on CT images.

PET/CT-based 3D multi-class semantic segmentation of ovarian cancer and the stability of the extracted radiomics features

Accurate segmentation of ovarian cancer (OC) lesions in PET/CT images is essential for effective disease management, yet manual segmentation for radiomics analysis is labor-intensive and time-consuming. This study introduces the application of a 3D U-Net deep learning model, leveraging advanced 3D networks, for multi-class semantic segmentation of OC in PET/CT images and assesses the stability of the extracted radiomics features. Utilizing a dataset of 3120 PET/CT images from 39 OC patients, the dataset was divided into training (70%), validation (15%), and test (15%) subsets to optimize and evaluate the model's performance. The 3D U-Net model, especially with a VGG16 backbone, achieved notable segmentation accuracy with a Dice score of 0.74, Precision of 0.76, and Recall of 0.78. Additionally, the study demonstrated high stability in radiomics features, with over 85% of PET and 84% of CT image features showing high intraclass correlation coefficients (ICCs > 0.8). These results underscore the potential of automated 3D U-Net-based segmentation to significantly enhance OC diagnosis and treatment planning. The reliability of the extracted radiomics features from automated segmentation supports its application in clinical decision-making and personalized medicine. This research marks a significant advancement in oncology diagnostics, providing a robust and efficient method for segmenting OC lesions in PET/CT images. By addressing the challenges of manual segmentation and demonstrating the effectiveness of 3D networks, this study contributes to the growing body of evidence supporting the application of artificial intelligence in improving diagnostic accuracy and patient outcomes in oncology.

Automatic segmentation of tumour and organs at risk in 3D MRI for cervical cancer radiation therapy with anatomical variations

AbstractCervical cancer is a common cancer in women globally, with treatment usually involving radiation therapy (RT). Accurate segmentation for the tumour site and organ-at-risks (OARs) could assist in the reduction of treatment side effects and improve treatment planning efficiency. Cervical cancer Magnetic Resonance Imaging (MRI) segmentation is challenging due to a limited amount of training data available and large inter- and intra- patient shape variation for OARs. The proposed Masked-Net consists of a masked encoder within the 3D U-Net to account for the large shape variation within the dataset, with additional dilated layers added to improve segmentation performance. A new loss function was introduced to consider the bounding box loss during training with the proposed Masked-Net. Transfer learning from a male pelvis MRI data with a similar field of view was included. The approaches were compared to the 3D U-Net which was widely used in MRI image segmentation. The data used consisted of 52 volumes obtained from 23 patients with stage IB to IVB cervical cancer across a maximum of 7 weeks of RT with manually contoured labels including the bladder, cervix, gross tumour volume, uterus and rectum. The model was trained and tested with a 5-fold cross validation. Outcomes were evaluated based on the Dice Similarity Coefficients (DSC), the Hausdorff Distance (HD) and the Mean Surface Distance (MSD). The proposed method accounted for the small dataset, large variations in OAR shape and tumour sizes with an average DSC, HD and MSD for all anatomical structures of 0.790, 30.19mm and 3.15mm respectively.

The application of multi-arc volumetric modulated arc therapy and fixed-field intensity modulated radiotherapy in the treatment of gynecologic cancer with large planning target volume

To investigate the dosimetry and delivery efficiency differences between multi-arc volumetric modulated arc therapy (VMAT) and fixed-field intensity modulated radiotherapy (IMRT) in the treatment of gynecological cancer with large planning target volume (PTV). Thirteen patients with gynecological cancer (9 cervical and 4 vulvar) with a PTV greater than 1600 cm

Lifetime radiation-induced sarcoma risk in patients subjected to IMRT or VMAT for uterine cervix carcinoma

This study was conducted to estimate the lifetime radiation-induced bone and soft tissue sarcoma risks from intensity modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT) for uterine cervix carcinoma. 13 cervical cancer patients were included. The bone and soft tissue structures were defined on patients' treatment planning computed tomography (CT) scans. Both CT-based IMRT and VMAT plans with 6 MV photons delivering 45 Gy to the target site were designed for each patient. The organ equivalent dose (OED) and the lifetime attributable risk (LAR) for developing bone or soft tissue sarcoma were estimated using treatment planning data and a non-linear mechanistic model. The estimation method did not consider the survival rates following radiotherapy and the use of brachytherapy treatments. The patient-specific OEDs of the bone structure from IMRT and VMAT were 2.33-2.83 and 2.34-2.82 Gy, respectively. The corresponding values for the soft tissue structure were 1.27-1.70 and 1.32-1.73 Gy. An insignificant difference was found between the patient-specific OEDs and the directly proportional sarcoma risks (bone: P = 0.07; soft tissue: P = 0.38). The LAR for the development of a bone sarcoma varied from 0.05 to 0.16% by the patient's age during irradiation and the applied treatment delivery technique. The corresponding LAR range for radiation-induced soft-tissue sarcoma was 0.08-0.27%. The above LARs resulted in a relative risk of more than 1.20 indicating that IMRT or VMAT may lead to a considerable risk increase of developing bone or soft tissue sarcoma exceeding 20% in respect to the current incidence of these malignancies in unexposed population.

Applicator-guided proton therapy versus multichannel brachytherapy for vaginal vault irradiation

To dosimetrically compare applicator-guided intensity-modulated proton therapy (IMPT) and multichannel brachytherapy (MC-BRT) for vaginal vault irradiation (VVI) with special focus on dose to organs at risk (OARs) and normal tissues. Ten patients with uterine confined endometrial cancer who received adjuvant vaginal cuff brachytherapy were included in this study. For each patient an additional IMPT treatment plan was created using the same computed tomography dataset and contours segmented for MC-BRT plans. Clinical target volume (CTV) was defined as the proximal 3.5 cm of the vagina including the entire thickness of vaginal wall. Planning target volume for IMPT plans was generated from the CTV with an addition isotropic 3 mm margin. OARs included rectum, bladder, sigmoid, small bowel and femoral heads. The prescribed dose was 21 Gy in 3 fractions. For simplicity, all doses were expressed in Gy and a constant relative biological effectiveness of 1.1 was used for IMPT plans. Plan comparison was performed using dose-volume histogram and treatment planning parameters. A significant improvement of the D98% coverage for CTV was reached by the applicator-guided IMPT plans (p < 0.01). IMPT also provided a dose reduction in all OARs except for femoral heads due to the lateral beam direction, especially significant reduction of V5Gy, D2cc, D0.1 cc, Dmean, V95% values for the rectum and Dmean, D0.1 cc to bladder, sigmoid, small bowel. Additionally, IMPT plans showed a significant reduction of integral dose to normal tissue with respect to MC-BRT (221.5 cGy.L vs. 653.6 cGy.L, p < 0.01). Applicator-guided IMPT has the potential for improving plan quality in VVI while maintaining the high conformity afforded by the state-of-the-art intracavitary brachytherapy.