DistilIQA: Distilling Vision Transformers for no-reference perceptual CT image quality assessment.

No-reference image quality assessment (IQA) is a critical step in medical image analysis, with the objective of predicting perceptual image quality without the need for a pristine reference image. The application of no-reference IQA to CT scans is valuable in providing an automated and objective approach to assessing scan quality, optimizing radiation dose, and improving overall healthcare efficiency. In this paper, we introduce DistilIQA, a novel distilled Vision Transformer network designed for no-reference CT image quality assessment. DistilIQA integrates convolutional operations and multi-head self-attention mechanisms by incorporating a powerful convolutional stem at the beginning of the traditional ViT network. Additionally, we present a two-step distillation methodology aimed at improving network performance and efficiency. In the initial step, a "teacher ensemble network" is constructed by training five vision Transformer networks using a five-fold division schema. In the second step, a "student network", comprising of a single Vision Transformer, is trained using the original labeled dataset and the predictions generated by the teacher network as new labels. DistilIQA is evaluated in the task of quality score prediction from low-dose chest CT scans obtained from the LDCT and Projection data of the Cancer Imaging Archive, along with low-dose abdominal CT images from the LDCTIQAC2023 Grand Challenge. Our results demonstrate DistilIQA's remarkable performance in both benchmarks, surpassing the capabilities of various CNNs and Transformer architectures. Moreover, our comprehensive experimental analysis demonstrates the effectiveness of incorporating convolutional operations within the ViT architecture and highlights the advantages of our distillation methodology.

CrossMoDA 2021 challenge: Benchmark of cross-modality domain adaptation techniques for vestibular schwannoma and cochlea segmentation.

Domain Adaptation (DA) has recently been of strong interest in the medical imaging community. While a large variety of DA techniques have been proposed for image segmentation, most of these techniques have been validated either on private datasets or on small publicly available datasets. Moreover, these datasets mostly addressed single-class problems. To tackle these limitations, the Cross-Modality Domain Adaptation (crossMoDA) challenge was organised in conjunction with the 24th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2021). CrossMoDA is the first large and multi-class benchmark for unsupervised cross-modality Domain Adaptation. The goal of the challenge is to segment two key brain structures involved in the follow-up and treatment planning of vestibular schwannoma (VS): the VS and the cochleas. Currently, the diagnosis and surveillance in patients with VS are commonly performed using contrast-enhanced T1 (ceT1) MR imaging. However, there is growing interest in using non-contrast imaging sequences such as high-resolution T2 (hrT2) imaging. For this reason, we established an unsupervised cross-modality segmentation benchmark. The training dataset provides annotated ceT1 scans (N=105) and unpaired non-annotated hrT2 scans (N=105). The aim was to automatically perform unilateral VS and bilateral cochlea segmentation on hrT2 scans as provided in the testing set (N=137). This problem is particularly challenging given the large intensity distribution gap across the modalities and the small volume of the structures. A total of 55 teams from 16 countries submitted predictions to the validation leaderboard. Among them, 16 teams from 9 different countries submitted their algorithm for the evaluation phase. The level of performance reached by the top-performing teams is strikingly high (best median Dice score - VS: 88.4%; Cochleas: 85.7%) and close to full supervision (median Dice score - VS: 92.5%; Cochleas: 87.7%). All top-performing methods made use of an image-to-image translation approach to transform the source-domain images into pseudo-target-domain images. A segmentation network was then trained using these generated images and the manual annotations provided for the source image.

Estimated Levator Ani Subtended Volume: A Predictive Biomarker for Surgical Outcomes Following Native Tissue Apical Repair.

OBJECTIVE: The aim of the study was to investigate the clinical utility of estimated levator ani subtended volume (eLASV) as a prospective preoperative biomarker for prediction of surgical outcomes. STUDY DESIGN: This is a prospective case-control pilot study. Patients were recruited and gave consent between January 2018 and December 2020. Surgical failure was defined by composite score. The eLASV was calculated for each patient based on a previously published algorithm. Descriptive statistics, Fisher exact test, log-binomial regression, area under a receiver operating characteristics, Bland-Altman plot, Lin coefficient, and κ coefficient were all performed for analysis. RESULTS: Fifty-one patients gave consent, 31 completed preoperative magnetic resonance imaging, 27 underwent surgery (uterosacral ligament suspension), and 19 followed up for 1-year examination. Five patients (26.3%) were defined as surgical failure with median eLASV volume of 57.0 (interquartile range, 50.1-66.2). Fourteen patients (73.7%) were defined as surgical success with median eLASV of 28.2 (interquartile range, 17.2-24.3). Eighty percent of the surgical failure group (4/5) had elevated volume of eLASV, where only 14.3% of the success group (2/14) had an elevated volume (P = 0.0173). No confounders were found and unadjusted log-binomial regression suggested that patients with a high eLASV were 8.7 (95% confidence interval, 1.2-61.9) times more likely to experience surgical failure compared with those with low eLASV. The c-statistic (area under a receiver operating characteristics) was high at 0.829 along with Lin concordance coefficient of 0.949 (95% confidence interval, 0.891-0.977) for continuous data between the 2 interrater observer teams. CONCLUSIONS: In this small prospective pilot study, patients with elevated eLASV on a preoperative pelvic magnetic resonance imaging were associated with an increased risk for surgical failure at 1 year regardless of age, body mass index, stage, or parity.CLINICAL TRIAL REGISTRATION:ClinicalTrials.gov, NCT03534830.

EMONAS-Net: Efficient multiobjective neural architecture search using surrogate-assisted evolutionary algorithm for 3D medical image segmentation.

Deep learning plays a critical role in medical image segmentation. Nevertheless, manually designing a neural network for a specific segmentation problem is a very difficult and time-consuming task due to the massive hyperparameter search space, long training time and large volumetric data. Therefore, most designed networks are highly complex, task specific and over-parametrized. Recently, multiobjective neural architecture search (NAS) methods have been proposed to automate the design of accurate and efficient segmentation architectures. However, they only search for either the micro- or macro-structure of the architecture, do not use the information produced during the optimization process to increase the efficiency of the search, or do not consider the volumetric nature of medical images. In this work, we present EMONAS-Net, an Efficient MultiObjective NAS framework for 3D medical image segmentation that optimizes both the segmentation accuracy and size of the network. EMONAS-Net has two key components, a novel search space that considers the configuration of the micro- and macro-structure of the architecture and a Surrogate-assisted Multiobjective Evolutionary based Algorithm (SaMEA algorithm) that efficiently searches for the best hyperparameter values. The SaMEA algorithm uses the information collected during the initial generations of the evolutionary process to identify the most promising subproblems and select the best performing hyperparameter values during mutation to improve the convergence speed. Furthermore, a Random Forest surrogate model is incorporated to accelerate the fitness evaluation of the candidate architectures. EMONAS-Net is tested on the tasks of prostate segmentation from the MICCAI PROMISE12 challenge, hippocampus segmentation from the Medical Segmentation Decathlon challenge, and cardiac segmentation from the MICCAI ACDC challenge. In all the benchmarks, the proposed framework finds architectures that perform better or comparable with competing state-of-the-art NAS methods while being considerably smaller and reducing the architecture search time by more than 50%.

AdaEn-Net: An ensemble of adaptive 2D-3D Fully Convolutional Networks for medical image segmentation.

Fully Convolutional Networks (FCNs) have emerged as powerful segmentation models but are usually designed manually, which requires extensive time and can result in large and complex architectures. There is a growing interest to automatically design efficient architectures that can accurately segment 3D medical images. However, most approaches either do not fully exploit volumetric information or do not optimize the model's size. To address these problems, we propose a self-adaptive 2D-3D ensemble of FCNs called AdaEn-Net for 3D medical image segmentation that incorporates volumetric data and adapts to a particular dataset by optimizing both the model's performance and size. The AdaEn-Net consists of a 2D FCN that extracts intra-slice information and a 3D FCN that exploits inter-slice information. The architecture and hyperparameters of the 2D and 3D architectures are found through a multiobjective evolutionary based algorithm that maximizes the expected segmentation accuracy and minimizes the number of parameters in the network. The main contribution of this work is a model that fully exploits volumetric information and automatically searches for a high-performing and efficient architecture. The AdaEn-Net was evaluated for prostate segmentation on the PROMISE12 Grand Challenge and for cardiac segmentation on the MICCAI ACDC challenge. In the first challenge, the AdaEn-Net ranks 9 out of 297 submissions and surpasses the performance of an automatically-generated segmentation network while producing an architecture with 13× fewer parameters. In the second challenge, the proposed model is ranked within the top 8 submissions and outperforms an architecture designed with reinforcement learning while having 1.25× fewer parameters.

Cost-effectiveness of a preoperative pelvic MRI in pelvic organ prolapse surgery.

OBJECTIVE: To investigate the cost-effectiveness of preoperative pelvic magnetic resonance imaging (MRI) in identifying women at high risk of surgical failure following apical repair for pelvic organ prolapse (POP). METHODS: A decision tree (TreeAgePro Healthcare software) was designed to compare outcomes and costs of screening with a pelvic MRI versus no screening. For the strategy with MRI, expected surgical outcomes were based on a calculated value of the estimated levator ani subtended volume (eLASV) from previously published work. For the alternative strategy of no MRI, estimates for surgical outcomes were obtained from the published literature. Costs for surgical procedures were estimated using the 2008-2014 National Inpatient Sample (NIS). A cost-effectiveness analysis from a third-party payer perspective was performed with the primary measure of effectiveness defined as avoidance of surgical failure. Deterministic and probabilistic sensitivity analyses were performed to assess how robust the calculated incremental cost-effectiveness ratio was to uncertainty in decision tree estimates and across a range of willingness-to-pay values. RESULTS: A preoperative MRI resulted in a 17% increased chance of successful initial surgery (87% vs. 70%) and a decreased risk of repeat surgery with an ICER of $2298 per avoided cost of surgical failure. When applied to annual expected women undergoing POP surgery, routine screening with preoperative pelvic MRI costs $90 million more, but could avoid 39,150 surgical failures. CONCLUSION: The use of routine preoperative pelvic MRI appears to be cost-effective when employed to identify women at high risk of surgical failure following apical repair for pelvic organ prolapse.

Automated contour tracking and trajectory classification of pelvic organs on dynamic MRI.

A method is presented to automatically track and segment pelvic organs on dynamic magnetic resonance imaging (MRI) followed by multiple-object trajectory classification to improve understanding of pelvic organ prolapse (POP). POP is a major health problem in women where pelvic floor organs fall from their normal position and bulge into the vagina. Dynamic MRI is presently used to analyze the organs' movements, providing complementary support for clinical examination. However, there is currently no automated or quantitative approach to measure the movement of the pelvic organs and their correlation with the severity of prolapse. In the proposed method, organs are first tracked and segmented using particle filters and [Formula: see text]-means clustering with prior information. Then, the trajectories of the pelvic organs are modeled using a coupled switched hidden Markov model to classify the severity of POP. Results demonstrate that the presented method can automatically track and segment pelvic organs with a Dice similarity index above 78% and Hausdorff distance of [Formula: see text] for 94 tested cases while demonstrating correlation between organ movement and POP. This work aims to enable automatic tracking and analysis of multiple deformable structures from images to improve understanding of medical disorders.

Automatic vertebra segmentation on dynamic magnetic resonance imaging.

The automatic extraction of the vertebra's shape from dynamic magnetic resonance imaging (MRI) could improve understanding of clinical conditions and their diagnosis. It is hypothesized that the shape of the sacral curve is related to the development of some gynecological conditions such as pelvic organ prolapse (POP). POP is a critical health condition for women and consists of pelvic organs dropping from their normal position. Dynamic MRI is used for assessing POP and to complement clinical examination. Studies have shown some evidence on the association between the shape of the sacral curve and the development of POP. However, the sacral curve is currently extracted manually limiting studies to small datasets and inconclusive evidence. A method composed of an adaptive shortest path algorithm that enhances edge detection and linking, and an improved curve fitting procedure is proposed to automate the identification and segmentation of the sacral curve on MRI. The proposed method uses predetermined pixels surrounding the sacral curve that are found through edge detection to decrease computation time compared to other model-based segmentation algorithms. Moreover, the proposed method is fully automatic and does not require user input or training. Experimental results show that the proposed method can accurately identify sacral curves for nearly 91% of dynamic MRI cases tested in this study. The proposed model is robust and can be used to effectively identify bone structures on MRI.

A web-based knowledge management system integrating Western and Traditional Chinese Medicine for relational medical diagnosis.

This article presents the design of a web-based knowledge management system as a training and research tool for the exploration of key relationships between Western and Traditional Chinese Medicine, in order to facilitate relational medical diagnosis integrating these mainstream healing modalities. The main goal of this system is to facilitate decision-making processes, while developing skills and creating new medical knowledge. Traditional Chinese Medicine can be considered as an ancient relational knowledge-based approach, focusing on balancing interrelated human functions to reach a healthy state. Western Medicine focuses on specialties and body systems and has achieved advanced methods to evaluate the impact of a health disorder on the body functions. Identifying key relationships between Traditional Chinese and Western Medicine opens new approaches for health care practices and can increase the understanding of human medical conditions. Our knowledge management system was designed from initial datasets of symptoms, known diagnosis and treatments, collected from both medicines. The datasets were subjected to process-oriented analysis, hierarchical knowledge representation and relational database interconnection. Web technology was implemented to develop a user-friendly interface, for easy navigation, training and research. Our system was prototyped with a case study on chronic prostatitis. This trial presented the system's capability for users to learn the correlation approach, connecting knowledge in Western and Traditional Chinese Medicine by querying the database, mapping validated medical information, accessing complementary information from official sites, and creating new knowledge as part of the learning process. By addressing the challenging tasks of data acquisition and modeling, organization, storage and transfer, the proposed web-based knowledge management system is presented as a tool for users in medical training and research to explore, learn and update relational information for the practice of integrated medical diagnosis. This proposal in education has the potential to enable further creation of medical knowledge from both Traditional Chinese and Western Medicine for improved care providing. The presented system positively improves the information visualization, learning process and knowledge sharing, for training and development of new skills for diagnosis and treatment, and a better understanding of medical diseases.

Automated Localization of Multiple Pelvic Bone Structures on MRI.

In this paper, we present a fully automated localization method for multiple pelvic bone structures on magnetic resonance images (MRI). Pelvic bone structures are at present identified manually on MRI to locate reference points for measurement and evaluation of pelvic organ prolapse (POP). Given that this is a time-consuming and subjective procedure, there is a need to localize pelvic bone structures automatically. However, bone structures are not easily differentiable from soft tissue on MRI as their pixel intensities tend to be very similar. In this paper, we present a model that combines support vector machines and nonlinear regression capturing global and local information to automatically identify the bounding boxes of bone structures on MRI. The model identifies the location of the pelvic bone structures by establishing the association between their relative locations and using local information such as texture features. Results show that the proposed method is able to locate the bone structures of interest accurately (dice similarity index >0.75) in 87-91% of the images. This research aims to enable accurate, consistent, and fully automated localization of bone structures on MRI to facilitate and improve the diagnosis of health conditions such as female POP.

Automated tracking, segmentation and trajectory classification of pelvic organs on dynamic MRI.

Pelvic organ prolapse is a major health problem in women where pelvic floor organs (bladder, uterus, small bowel, and rectum) fall from their normal position and bulge into the vagina. Dynamic Magnetic Resonance Imaging (DMRI) is presently used to analyze the organs' movements from rest to maximum strain providing complementary support for diagnosis. However, there is currently no automated or quantitative approach to measure the movement of the pelvic organs and their correlation with the severity of prolapse. In this paper, a two-stage method is presented to automatically track and segment pelvic organs on DMRI followed by a multiple-object trajectory classification method to improve the diagnosis of pelvic organ prolapse. Organs are first tracked using particle filters and K-means clustering with prior information. Then, they are segmented using the convex hull of the cluster of particles. Finally, the trajectories of the pelvic organs are modeled using a new Coupled Switched Hidden Markov Model (CSHMM) to classify the severity of pelvic organ prolapse. The tracking and segmentation results are validated using Dice Similarity Index (DSI) whereas the classification results are compared with two manual clinical measurements. Results demonstrate that the presented method is able to automatically track and segment pelvic organs with a DSI above 82% for 26 out of 46 cases and DSI above 75% for all 46 tested cases. The accuracy of the trajectory classification model is also better than current manual measurements.

MRI-based segmentation of pubic bone for evaluation of pelvic organ prolapse.

Pelvic organ prolapse (POP) is a major women's health problem. Its diagnosis through magnetic resonance imaging (MRI) has become popular due to current inaccuracies of clinical examination. The diagnosis of POP on MRI consists of identifying reference points on pelvic bone structures for measurement and evaluation. However, it is currently performed manually, making it a time-consuming and subjective procedure. We present a new segmentation approach for automating pelvic bone point identification on MRI. It consists of a multistage mechanism based on texture-based block classification, leak detection, and prior shape information. Texture-based block classification and clustering analysis using K-means algorithm are integrated to generate the initial bone segmentation and to identify leak areas. Prior shape information is incorporated to obtain the final bone segmentation. Then, the reference points are identified using morphological skeleton operation. Results demonstrate that the proposed method achieves higher bone segmentation accuracy compared to other segmentation methods. The proposed method can also automatically identify reference points faster and with more consistency compared with the manually identified point process by experts. This research aims to enable faster and consistent pelvic measurements on MRI to facilitate and improve the diagnosis of female POP.

Fully automated localization of multiple pelvic bone structures on MRI.

In this paper, we present a fully automated localization method for multiple pelvic bone structures on magnetic resonance images (MRI). Pelvic bone structures are currently identified manually on MRI to identify reference points for measurement and evaluation of pelvic organ prolapse (POP). Given that this is a time-consuming and subjective procedure, there is a need to localize pelvic bone structures without any user interaction. However, bone structures are not easily differentiable from soft tissue on MRI as their pixel intensities tend to be very similar. In this research, we present a model that automatically identifies the bounding boxes of the bone structures on MRI using support vector machines (SVM) based classification and non-linear regression model that captures global and local information. Based on the relative locations of pelvic bones and organs, and local information such as texture features, the model identifies the location of the pelvic bone structures by establishing the association between their locations. Results show that the proposed method is able to locate the bone structures of interest accurately. The pubic bone, sacral promontory, and coccyx were correctly detected (DSI > 0.75) in 92%, 90%, and 88% of the testing images. This research aims to enable accurate, consistent and fully automated identification of pelvic bone structures on MRI to facilitate and improve the diagnosis of female pelvic organ prolapse.

Synergetic effects of doxycycline-loaded chitosan nanoparticles for improving drug delivery and efficacy.

INTRODUCTION: Doxycycline, a broad-spectrum antibiotic, is the most commonly prescribed antibiotic worldwide for treating infectious diseases. It may be delivered orally or intravenously but can lead to gastrointestinal irritation and local inflammation. For treatment of uterine infections, transcervical administration of doxycycline encapsulated in nanoparticles made of biodegradable chitosan may improve sustained delivery of the drug, thereby minimizing adverse effects and improving drug efficacy. METHODS AND MATERIALS: As a first step toward assessing this potential, we used an ionic gelation method to synthesize blank and doxycycline-loaded chitosan nanoparticles (DCNPs), which we then characterized in terms of several properties relevant to clinical efficacy: particle size, shape, encapsulation efficiency, antibacterial activity, and in vitro cytotoxicity. Two particle formulations were examined, with one (named DCNP6) containing approximately 1.5 times the crosslinker concentration of the other (DCNP4). RESULTS: The two formulations produced spherically shaped drug-loaded nanoparticles. The spheres ranged in size from 30 to 220 nm diameter for DCNP4 and 200 to 320 nm diameter for DCNP6. Average encapsulation yield was 53% for DCNP4 and 56% for DCNP6. In terms of drug release, both formulations showed a burst effect within the first 4 to 5 hours, followed by a slow, sustained release for the remainder of the 24-hour monitoring period. The in vitro antibacterial activity against Escherichia coli was high, with both formulations achieving more than 90% inhibition of 4-hour bacterial growth. Cytotoxic effects of the DCNPs on normal human ovarian surface epithelial cells were significantly lower than those of unencapsulated doxycycline. After 5 days, cultures exposed to the unencapsulated antibiotic showed a 61% decrease in cell viability, while cultures exposed to the DCNPs exhibited less than a 10% decrease. CONCLUSION: These laboratory results suggest that DCNPs show preliminary promise for possible eventual use in transcervical drug delivery and improved efficacy in the treatment of bacterial uterine infections.