Cell classification with worse-case boosting for intelligent cervical cancer screening.

Cell classification underpins intelligent cervical cancer screening, a cytology examination that effectively decreases both the morbidity and mortality of cervical cancer. This task, however, is rather challenging, mainly due to the difficulty of collecting a training dataset representative sufficiently of the unseen test data, as there are wide variations of cells' appearance and shape at different cancerous statuses. This difficulty makes the classifier, though trained properly, often classify wrongly for cells that are underrepresented by the training dataset, eventually leading to a wrong screening result. To address it, we propose a new learning algorithm, called worse-case boosting, for classifiers effectively learning from under-representative datasets in cervical cell classification. The key idea is to learn more from worse-case data for which the classifier has a larger gradient norm compared to other training data, so these data are more likely to correspond to underrepresented data, by dynamically assigning them more training iterations and larger loss weights for boosting the generalizability of the classifier on underrepresented data. We achieve this idea by sampling worse-case data per the gradient norm information and then enhancing their loss values to update the classifier. We demonstrate the effectiveness of this new learning algorithm on two publicly available cervical cell classification datasets (the two largest ones to the best of our knowledge), and positive results (4% accuracy improvement) yield in the extensive experiments. The source codes are available at: https://github.com/YouyiSong/Worse-Case-Boosting.

Intra-Patient Lung CT Registration through Large Deformation Decomposition and Attention-Guided Refinement.

Deformable lung CT image registration is an essential task for computer-assisted interventions and other clinical applications, especially when organ motion is involved. While deep-learning-based image registration methods have recently achieved promising results by inferring deformation fields in an end-to-end manner, large and irregular deformations caused by organ motion still pose a significant challenge. In this paper, we present a method for registering lung CT images that is tailored to the specific patient being imaged. To address the challenge of large deformations between the source and target images, we break the deformation down into multiple continuous intermediate fields. These fields are then combined to create a spatio-temporal motion field. We further refine this field using a self-attention layer that aggregates information along motion trajectories. By leveraging temporal information from a respiratory cycle, our proposed methods can generate intermediate images that facilitate image-guided tumor tracking. We evaluated our approach extensively on a public dataset, and our numerical and visual results demonstrate the effectiveness of the proposed method.

Data Discernment for Affordable Training in Medical Image Segmentation.

Collecting sufficient high-quality training data for deep neural networks is often expensive or even unaffordable in medical image segmentation tasks. We thus propose to train the network by using external data that can be collected in a cheaper way, e.g., crowd-sourcing. We show that by data discernment, the network is able to mine valuable knowledge from external data, even though the data distribution is very different from that of the original (internal) data. We discern the external data by learning an importance weight for each of them, with the goal to enhance the contribution of informative external data to network updating, while suppressing the data that are 'useless' or even 'harmful'. An iterative algorithm that alternatively estimates the importance weight and updates the network is developed by formulating the data discernment as a constrained nonlinear programming problem. It estimates the importance weight according to the distribution discrepancy between the external data and the internal dataset, and imposes a constraint to drive the network to learn more effectively, compared with the network without using the external data. We evaluate the proposed algorithm on two tasks: abdominal CT image and cervical smear image segmentation, using totally 6 publicly available datasets. The effectiveness of the algorithm is demonstrated by extensive experiments. Source codes are available at: https://github.com/YouyiSong/Data-Discernment.

MDGF-MCEC: a multi-view dual attention embedding model with cooperative ensemble learning for CircRNA-disease association prediction.

Circular RNA (circRNA) is closely involved in physiological and pathological processes of many diseases. Discovering the associations between circRNAs and diseases is of great significance. Due to the high-cost to verify the circRNA-disease associations by wet-lab experiments, computational approaches for predicting the associations become a promising research direction. In this paper, we propose a method, MDGF-MCEC, based on multi-view dual attention graph convolution network (GCN) with cooperative ensemble learning to predict circRNA-disease associations. First, MDGF-MCEC constructs two disease relation graphs and two circRNA relation graphs based on different similarities. Then, the relation graphs are fed into a multi-view GCN for representation learning. In order to learn high discriminative features, a dual-attention mechanism is introduced to adjust the contribution weights, at both channel level and spatial level, of different features. Based on the learned embedding features of diseases and circRNAs, nine different feature combinations between diseases and circRNAs are treated as new multi-view data. Finally, we construct a multi-view cooperative ensemble classifier to predict the associations between circRNAs and diseases. Experiments conducted on the CircR2Disease database demonstrate that the proposed MDGF-MCEC model achieves a high area under curve of 0.9744 and outperforms the state-of-the-art methods. Promising results are also obtained from experiments on the circ2Disease and circRNADisease databases. Furthermore, the predicted associated circRNAs for hepatocellular carcinoma and gastric cancer are supported by the literature. The code and dataset of this study are available at https://github.com/ABard0/MDGF-MCEC.

Deep Cross-Output Knowledge Transfer Using Stacked-Structure Least-Squares Support Vector Machines.

This article presents a new deep cross-output knowledge transfer approach based on least-squares support vector machines, called DCOT-LS-SVMs. Its aim is to improve the generalizability of least-squares support vector machines (LS-SVMs) while avoiding the complicated parameter tuning process that occurs in many kernel machines. The proposed approach has two significant characteristics: 1) DCOT-LS-SVMs is inspired by a stacked hierarchical architecture that combines several layer-by-layer LS-SVMs modules. The module that forms the higher layer has additional input features that consider the predictions from all previous modules and 2) cross-output knowledge transfer is used to leverage knowledge from the predictions of the previous module to improve the learning process in the current module. With this approach, the model's parameters, such as a tradeoff parameter C and a kernel width δ , can be randomly assigned to each module in order to greatly simplify the learning process. Moreover, DCOT-LS-SVMs is able to autonomously and quickly decide the extent of the cross-output knowledge transfer between adjacent modules through a fast leave-one-out cross-validation strategy. In addition, we present an imbalanced version of DCOT-LS-SVMs, called IDCOT-LS-SVMs, given that imbalanced datasets are common in real-world scenarios. The effectiveness of the proposed approaches is demonstrated through a comparison with five comparative methods on UCI datasets and with a case study on the diagnosis of prostate cancer.

Enhancement of prostate cancer diagnosis by machine learning techniques: an algorithm development and validation study.

BACKGROUND: To investigate the value of machine learning(ML) in enhancing prostate cancer(PCa) diagnosis. METHODS: Consecutive systematic prostate biopsies performed from Jan 2003-June 2017 were used as the training cohort, and prospective biopsies performed from July 2017-November 2019 were used as validation cohort. Men were included if PSA was 0.4-50 ng/mL, and information of digital rectal examination (DRE), Transrectal ultrasound(TRUS) prostate volume, TRUS abnormality were known. Clinically significant PCa(csPCa) was defined as Gleason 3 + 4 or above cancers. Area-under-curve (AUC) of receiver-operating characteristics (ROC) was compared between PSA, PSA density, European Randomized Study of Screening for Prostate Cancer (ERSPC) risk calculator (ERSPC-RC), and various ML techniques using PSA, DRE and TRUS information. ML techniques used included XGBoost, LightGBM, Catboost, Support vector machine (SVM), Logistic regression (LR), and Random Forest (RF), where cost sensitive learning was applied. RESULTS: Training and validation cohorts included 3881 and 778 consecutive men, respectively. RF model performed better than other ML techniques and PSA, PSA density and ERSPC-RC for prediction of PCa or csPCa in the validation cohort. In csPCa prediction, AUC of PSA, PSA density, ERSPC-RC and RF was 0.71, 0.80, 0.83 and 0.88 respectively. At 90-95% sensitivity for csPCa, RF model achieved a negative predictive value (NPV) of 97.5-98.0% and avoided 38.3-52.2% unnecessary biopsies. Decision curve analyses (DCA) showed RF model provided net clinical benefit over PSA, PSA density and ERSPC-RC. CONCLUSION: By using the same clinical parameters, ML techniques performed better than ERSPC-RC or PSA density in csPCa predictions, and could avoid up to 50% unnecessary biopsies.

Extended Kalman filter for online soft tissue characterization based on Hunt-Crossley contact model.

Real-time soft tissue characterization is significant to robotic assisted minimally invasive surgery for achieving precise haptic control of robotic surgical tasks and providing realistic force feedback to the operator. This paper presents a nonlinear methodology for online soft tissue characterization. An extended Kalman filter (EKF) is developed based on dynamic linearization of the nonlinear H-C contact model in terms of system state for online characterization of soft tissue parameters. To handle the resultant linearization modelling error, an innovation orthogonal EKF is further developed by incorporating an adaptive factor in the EKF filtering to adaptively adjust the innovation covariance according to the principle of innovation orthogonality. Simulation and experimental results as well as comparison analysis demonstrate that the proposed methodology can effectively characterize soft tissue parameters, leading to dramatically improved accuracy comparing to recursive least square estimation. Further, the proposed methodology also requires a smaller computational load and can achieve the real-time performance for soft tissue characterization.

Extended Kalman Filter Nonlinear Finite Element Method for Nonlinear Soft Tissue Deformation.

BACKGROUND AND OBJECTIVE: Soft tissue modelling is crucial to surgery simulation. This paper introduces an innovative approach to realistic simulation of nonlinear deformation behaviours of biological soft tissues in real time. METHODS: This approach combines the traditional nonlinear finite-element method (NFEM) and nonlinear Kalman filtering to address both physical fidelity and real-time performance for soft tissue modelling. It defines tissue mechanical deformation as a nonlinear filtering process for dynamic estimation of nonlinear deformation behaviours of biological tissues. Tissue mechanical deformation is discretized in space using NFEM in accordance with nonlinear elastic theory and in time using the central difference scheme to establish the nonlinear state-space models for dynamic filtering. RESULTS: An extended Kalman filter is established to dynamically estimate nonlinear mechanical deformation of biological tissues. Interactive deformation of biological soft tissues with haptic feedback is accomplished as well for surgery simulation. CONCLUSIONS: The proposed approach conquers the NFEM limitation of step computation but without trading off the modelling accuracy. It not only has a similar level of accuracy as NFEM, but also meets the real-time requirement for soft tissue modelling.

Segmentation of Overlapping Cytoplasm in Cervical Smear Images via Adaptive Shape Priors Extracted From Contour Fragments.

We present a novel approach for segmenting overlapping cytoplasm of cells in cervical smear images by leveraging the adaptive shape priors extracted from cytoplasm's contour fragments and shape statistics. The main challenge of this task is that many occluded boundaries in cytoplasm clumps are extremely difficult to be identified and, sometimes, even visually indistinguishable. Given a clump where multiple cytoplasms overlap, our method starts by cutting its contour into a set of contour fragments. We then locate the corresponding contour fragments of each cytoplasm by a grouping process. For each cytoplasm, according to the grouped fragments and a set of known shape references, we construct its shape and, then, connect the fragments to form a closed contour as the segmentation result, which is explicitly constrained by the constructed shape. We further integrate the intensity and curvature information, which is complementary to the shape priors extracted from contour fragments, into our framework to improve the segmentation accuracy. We propose to iteratively conduct fragments grouping, shape constructing, and fragments connecting for progressively refining the shape priors and improving the segmentation results. We extensively evaluate the effectiveness of our method on two typical cervical smear datasets. The experimental results demonstrate that our approach is highly effective and consistently outperforms the state-of-the-art approaches. The proposed method is general enough to be applied to other similar microscopic image segmentation tasks, where heavily overlapped objects exist.

Diagnosis of prostate cancer in a Chinese population by using machine learning methods.

An early diagnosis of prostate cancer (PC) is key for the successful treatment. Although invasive prostate biopsies can provide a definitive diagnosis, the number of biopsies should be reduced to avoid side effects and risks especially for the men with the low risk of cancer. Therefore, an accurate model is in need to predict PC with the aim of reducing unnecessary biopsies. In this study, we developed predictive models using four machine learning methods including Support Vector Machine (SVM), Least Squares Support Vector Machine (LS-SVM), Artificial Neural Network (ANN) and Random Forest (RF) to detect PC cases using available prebiopsy information. The models were constructed and evaluated on a cohort of 1625 Chinese men with prostate biopsies from Hong Kong hospital. All the models have the excellent performances in detecting significant PC cases, with ANN achieving the highest accuracy of 0.9527 and the AUC value of 0.9755. RF outperformed the other three methods in classifying benign, significant and insignificant PC cases, with an accuracy of 0.9741 and a F1 score of 0.8290.

A Computer-Based Method for Teaching Catheter-Access Hemodialysis Management.

Patients undergoing hemodialysis are highly susceptible to infections, which could lead to morbidity and mortality. One of the major sources of infections stems from the mishandling of hemodialysis access sites. Although healthcare workers receive training on how to aseptically handle hemodialysis catheters, the increasing number of blood infections associated with dialysis suggests that the conventional approach to training may not be sufficient to ensure a clear understanding of the necessary knowledge and skills. With advancements in digital technology, computer-assisted learning has been gaining popularity as an approach to teaching clinical skills. The purpose of this study was to evaluate the effectiveness of a computer-based training system developed to teach healthcare workers catheter-access hemodialysis management. Forty nurses were recruited and randomly assigned into two groups: the control group, which received conventional training only; and the experimental group, which received both conventional and computer-based training. A knowledge test and a skills competence test were administered to both groups before and after the intervention to evaluate their performance. The results show that the performance of the nurses in the experimental group was significantly better than that in the control group, indicating that the proposed training system is an effective tool for supplementing the learning of catheter-access hemodialysis management.

Enhancing training performance for brain-computer interface with object-directed 3D visual guidance.

PURPOSE: The accuracy of the classification of user intentions is essential for motor imagery (MI)-based brain-computer interface (BCI). Effective and appropriate training for users could help us produce the high reliability of mind decision making related with MI tasks. In this study, we aimed to investigate the effects of visual guidance on the classification performance of MI-based BCI. METHODS: In this study, leveraging both the single-subject and the multi-subject BCI paradigms, we train and classify MI tasks with three different scenarios in a 3D virtual environment, including non-object-directed scenario, static-object-directed scenario, and dynamic object-directed scenario. Subjects are required to imagine left-hand or right-hand movement with the visual guidance. RESULTS: We demonstrate that the classification performances of left-hand and right-hand MI task have differences on these three scenarios, and confirm that both static-object-directed and dynamic object-directed scenarios could provide better classification accuracy than the non-object-directed case. We further indicate that both static-object-directed and dynamic object-directed scenarios could shorten the response time as well as be suitable applied in the case of small training data. In addition, experiment results demonstrate that the multi-subject BCI paradigm could improve the classification performance comparing with the single-subject paradigm. These results suggest that it is possible to improve the classification performance with the appropriate visual guidance and better BCI paradigm. CONCLUSION: We believe that our findings would have the potential for improving classification performance of MI-based BCI and being applied in the practical applications.

Prediction of mortality after radical cystectomy for bladder cancer by machine learning techniques.

Bladder cancer is a common cancer in genitourinary malignancy. For muscle invasive bladder cancer, surgical removal of the bladder, i.e. radical cystectomy, is in general the definitive treatment which, unfortunately, carries significant morbidities and mortalities. Accurate prediction of the mortality of radical cystectomy is therefore needed. Statistical methods have conventionally been used for this purpose, despite the complex interactions of high-dimensional medical data. Machine learning has emerged as a promising technique for handling high-dimensional data, with increasing application in clinical decision support, e.g. cancer prediction and prognosis. Its ability to reveal the hidden nonlinear interactions and interpretable rules between dependent and independent variables is favorable for constructing models of effective generalization performance. In this paper, seven machine learning methods are utilized to predict the 5-year mortality of radical cystectomy, including back-propagation neural network (BPN), radial basis function (RBFN), extreme learning machine (ELM), regularized ELM (RELM), support vector machine (SVM), naive Bayes (NB) classifier and k-nearest neighbour (KNN), on a clinicopathological dataset of 117 patients of the urology unit of a hospital in Hong Kong. The experimental results indicate that RELM achieved the highest average prediction accuracy of 0.8 at a fast learning speed. The research findings demonstrate the potential of applying machine learning techniques to support clinical decision making.

Using interactive computer simulation for teaching the proper use of personal protective equipment.

The use of personal protective equipment is one of the basic infection control precautions in health care. The effectiveness of personal protective equipment is highly dependent on adequate staff training. In this project, a computer simulation program, as a supplement to conventional training approaches, was developed to facilitate the learning of the proper use of personal protective equipment. The simulation program was a Web-based interactive software with user-friendly graphical interface for users to practice the use of personal protective equipment usage via drag-and-drop metaphors and respond to questions online. The effectiveness of the computer simulation software was investigated by a controlled study. Fifty healthcare workers were randomly assigned into two groups: one received conventional personal protective equipment training only (control group), whereas the other also received the same conventional training but followed by using the developed simulation program for self-learning (experimental group). Their performance was assessed by personal protective equipment donning and doffing evaluation before and after the training. The results showed that the computer simulation program is able to improve the healthcare workers' understanding and competence in using personal protective equipment.

Learning blood management in orthopedic surgery through gameplay.

Orthopedic surgery treats the musculoskeletal system, in which bleeding is common and can be fatal. To help train future surgeons in this complex practice, researchers designed and implemented a serious game for learning orthopedic surgery. The game focuses on teaching trainees blood management skills, which are critical for safe operations. Using state-of-the-art graphics technologies, the game provides an interactive and realistic virtual environment. It also integrates game elements, including task-oriented and time-attack scenarios, bonuses, game levels, and performance evaluation tools. To study the system's effect, the researchers conducted experiments on player completion time and off-target contacts to test their learning of psychomotor skills in blood management.

Collaborative simulation of soft-tissue deformation for virtual surgery applications.

A key challenge of collaborative surgical simulation is to maintain a high level of state consistency among the distributed users under the limitation of network transmission capacity. In this paper, a framework integrating a scalable deformable model and an extensible communication protocol is proposed to meet this challenge. The parameters of the deformable model are obtained by making reference to the biomechanical properties of human soft tissues. Efficient collaboration is achieved by developing the communication protocol and implementing a series of network management approaches, including service management, computation policies, coupling control, token control and availability mechanism. A prototype has been developed by using the client-server network architecture. Experimental results demonstrate that this framework can support collaborative surgical simulation with acceptable network latencies.

A virtual training simulator for learning cataract surgery with phacoemulsification.

This paper presents the development of a low-cost cataract surgery simulator for trainees to practise phacoemulsification procedures with computer-generated models in virtual environments. It focuses on the training of cornea incision, capsulorrhexis and phaco-sculpting, which are simulated interactively with computationally efficient algorithms developed for tissue deformation, surface cutting and volume sculpting. Intuitive two-handed human-computer interactions are achieved with six degrees-of-freedom haptic devices. Performance of trainees on manual dexterity is recorded with quantifiable metrics. The proposed virtual-reality system has the potential to serve as an alternative training tool to supplement conventional cataract surgery education.

A framework using cluster-based hybrid network architecture for collaborative virtual surgery.

Research on collaborative virtual environments (CVEs) opens the opportunity for simulating the cooperative work in surgical operations. It is however a challenging task to implement a high performance collaborative surgical simulation system because of the difficulty in maintaining state consistency with minimum network latencies, especially when sophisticated deformable models and haptics are involved. In this paper, an integrated framework using cluster-based hybrid network architecture is proposed to support collaborative virtual surgery. Multicast transmission is employed to transmit updated information among participants in order to reduce network latencies, while system consistency is maintained by an administrative server. Reliable multicast is implemented using distributed message acknowledgment based on cluster cooperation and sliding window technique. The robustness of the framework is guaranteed by the failure detection chain which enables smooth transition when participants join and leave the collaboration, including normal and involuntary leaving. Communication overhead is further reduced by implementing a number of management approaches such as computational policies and collaborative mechanisms. The feasibility of the proposed framework is demonstrated by successfully extending an existing standalone orthopedic surgery trainer into a collaborative simulation system. A series of experiments have been conducted to evaluate the system performance. The results demonstrate that the proposed framework is capable of supporting collaborative surgical simulation.

Interactive cutting simulation with adaptive refinements using digital logic design analogy.

A major requirement for surgical simulation is to allow virtual tissue cutting. This paper presents a scalable and adaptive cutting technique based on a mass-spring mesh. By the analogy of digital logic design, an arbitrary incision is modeled systematically by translating the cutting process into a state diagram. Subdivision of mesh elements is driven by the state transitions. Node redistribution, local re-meshing and deformation are applied to refine the subdivided mesh.