Automatic 3D reconstruction of vertebrae from orthogonal bi-planar radiographs.

When conducting spine-related diagnosis and surgery, the three-dimensional (3D) upright posture of the spine under natural weight bearing is of significant clinical value for physicians to analyze the force on the spine. However, existing medical imaging technologies cannot meet current requirements of medical service. On the one hand, the mainstream 3D volumetric imaging modalities (e.g. CT and MRI) require patients to lie down during the imaging process. On the other hand, the imaging modalities conducted in an upright posture (e.g. radiograph) can only realize 2D projections, which lose the valid information of spinal anatomy and curvature. Developments of deep learning-based 3D reconstruction methods bring potential to overcome the limitations of the existing medical imaging technologies. To deal with the limitations of current medical imaging technologies as is described above, in this paper, we propose a novel deep learning framework, ReVerteR, which can realize automatic 3D Reconstruction of Vertebrae from orthogonal bi-planar Radiographs. With the utilization of self-attention mechanism and specially designed loss function combining Dice, Hausdorff, Focal, and MSE, ReVerteR can alleviate the sample-imbalance problem during the reconstruction process and realize the fusion of the centroid annotation and the focused vertebra. Furthermore, aiming at automatic and customized 3D spinal reconstruction in real-world scenarios, we extend ReVerteR to a clinical deployment-oriented framework, and develop an interactive interface with all functions in the framework integrated so as to enhance human-computer interaction during clinical decision-making. Extensive experiments and visualization conducted on our constructed datasets based on two benchmark datasets of spinal CT, VerSe 2019 and VerSe 2020, demonstrate the effectiveness of our proposed ReVerteR. In this paper, we propose an automatic 3D reconstruction method of vertebrae based on orthogonal bi-planar radiographs. With the 3D upright posture of the spine under natural weight bearing effectively constructed, our proposed method is expected to better support doctors make clinical decision during spine-related diagnosis and surgery.

Automatic extraction of facial median sagittal plane for patients with asymmetry based on the EDMA alignment algorithm.

BACKGROUND: We aimed to establish a novel method for automatically constructing three-dimensional (3D) median sagittal plane (MSP) for mandibular deviation patients, which can increase the efficiency of aesthetic evaluating treatment progress. We developed a Euclidean weighted Procrustes analysis (EWPA) algorithm for extracting 3D facial MSP based on the Euclidean distance matrix analysis, automatically assigning weight to facial anatomical landmarks. METHODS: Forty patients with mandibular deviation were recruited, and the Procrustes analysis (PA) algorithm based on the original mirror alignment and EWPA algorithm developed in this study were used to construct the MSP of each facial model of the patient as experimental groups 1 and 2, respectively. The expert-defined regional iterative closest point algorithm was used to construct the MSP as the reference group. The angle errors of the two experimental groups were compared to those of the reference group to evaluate their clinical suitability. RESULTS: The angle errors of the MSP constructed by the two EWPA and PA algorithms for the 40 patients were 1.39 ± 0.85°, 1.39 ± 0.78°, and 1.91 ± 0.80°, respectively. The two EWPA algorithms performed best in patients with moderate facial asymmetry, and in patients with severe facial asymmetry, the angle error was below 2°, which was a significant improvement over the PA algorithm. CONCLUSIONS: The clinical application of the EWPA algorithm based on 3D facial morphological analysis for constructing a 3D facial MSP for patients with mandibular deviated facial asymmetry deformity showed a significant improvement over the conventional PA algorithm and achieved the effect of a dental clinical expert-level diagnostic strategy.

Post-pandemic healthcare for COVID-19 vaccine: Tissue-aware diagnosis of cervical lymphadenopathy via multi-modal ultrasound semantic segmentation.

With the widespread deployment of COVID-19 vaccines all around the world, billions of people have benefited from the vaccination and thereby avoiding infection. However, huge amount of clinical cases revealed diverse side effects of COVID-19 vaccines, among which cervical lymphadenopathy is one of the most frequent local reactions. Therefore, rapid detection of cervical lymph node (LN) is essential in terms of vaccine recipients' healthcare and avoidance of misdiagnosis in the post-pandemic era. This paper focuses on a novel deep learning-based framework for the rapid diagnosis of cervical lymphadenopathy towards COVID-19 vaccine recipients. Existing deep learning-based computer-aided diagnosis (CAD) methods for cervical LN enlargement mostly only depend on single modal images, e.g., grayscale ultrasound (US), color Doppler ultrasound, and CT, while failing to effectively integrate information from the multi-source medical images. Meanwhile, both the surrounding tissue objects of the cervical LNs and different regions inside the cervical LNs may imply valuable diagnostic knowledge which is pending for mining. In this paper, we propose an Tissue-Aware Cervical Lymph Node Diagnosis method (TACLND) via multi-modal ultrasound semantic segmentation. The method effectively integrates grayscale and color Doppler US images and realizes a pixel-level localization of different tissue objects, i.e., lymph, muscle, and blood vessels. With inter-tissue and intra-tissue attention mechanisms applied, our proposed method can enhance the implicit tissue-level diagnostic knowledge in both spatial and channel dimension, and realize diagnosis of cervical LN with normal, benign or malignant state. Extensive experiments conducted on our collected cervical LN US dataset demonstrate the effectiveness of our methods on both tissue detection and cervical lymphadenopathy diagnosis. Therefore, our proposed framework can guarantee efficient diagnosis for the vaccine recipients' cervical LN, and assist doctors to discriminate between COVID-related reactive lymphadenopathy and metastatic lymphadenopathy.

Topology-enhanced molecular graph representation for anti-breast cancer drug selection.

BACKGROUND: Breast cancer is currently one of the cancers with a higher mortality rate in the world. The biological research on anti-breast cancer drugs focuses on the activity of estrogen receptors alpha (ER[Formula: see text]), the pharmacokinetic properties and the safety of the compounds, which, however, is an expensive and time-consuming process. Developments of deep learning bring potential to efficiently facilitate the candidate drug selection against breast cancer. METHODS: In this paper, we propose an Anti-Breast Cancer Drug selection method utilizing Gated Graph Neural Networks (ABCD-GGNN) to topologically enhance the molecular representation of candidate drugs. By constructing atom-level graphs through atomic descriptors for each distinct compound, ABCD-GGNN can topologically learn both the implicit structure and substructure characteristics of a candidate drug and then integrate the representation with explicit discrete molecular descriptors to generate a molecule-level representation. As a result, the representation of ABCD-GGNN can inductively predict the ER[Formula: see text], the pharmacokinetic properties and the safety of each candidate drug. Finally, we design a ranking operator whose inputs are the predicted properties so as to statistically select the appropriate drugs against breast cancer. RESULTS: Extensive experiments conducted on our collected anti-breast cancer candidate drug dataset demonstrate that our proposed method outperform all the other representative methods in the tasks of predicting ER[Formula: see text], and the pharmacokinetic properties and safety of the compounds. Extended result analysis demonstrates the efficiency and biological rationality of the operator we design to calculate the candidate drug ranking from the predicted properties. CONCLUSION: In this paper, we propose the ABCD-GGNN representation method to efficiently integrate the topological structure and substructure features of the molecules with the discrete molecular descriptors. With a ranking operator applied, the predicted properties efficiently facilitate the candidate drug selection against breast cancer.

Predicting Organization Performance Changes: A Sequential Data-Based Framework.

The business environment is increasingly uncertain due to the rapid development of disruptive information technologies, the changing global economy, and the COVID-19 pandemic. This brings great uncertainties to investors to predict the performance changes and risks of companies. This research proposes a sequential data-based framework that aggregates data from multiple sources including both structured and unstructured data to predict the performance changes. It leverages data generated from the early risk warning system in China stock market to measure and predict organization performance changes based on the risk warning status changes of public companies. Different from the models in existing literature that focus on the prediction of risk warning of companies, our framework predicts a portfolio of organization performance changes, including business decline and recovery, thus helping investors to not only predict public company risks, but also discover investment opportunities. By incorporating sequential data, our framework achieves 92.3% macro-F1 value on real-world data from listed companies in China, outperforming other static models.

A mathematical algorithm of the facial symmetry plane: Application to mandibular deformity 3D facial data.

The three-dimensional (3D) symmetry reference plane (SRP) is the premise and basis of 3D facial symmetry analysis. Currently, most methods for extracting the SRP are based on anatomical landmarks measured manually using a digital 3D facial model. However, as different clinicians have varying definitions of landmarks, establishing common methods suitable for different types of facial asymmetry remains challenging. The present study aimed to investigate and evaluate a novel mathematical algorithm based on power function weighted Procrustes analysis (PWPA) to determine 3D facial SRPs for patients with mandibular deviation. From 30 patients with mandibular deviation, 3D facial SRPs were determined using both our PWPA algorithms (two functions) and the traditional PA algorithm (experimental groups). A reference plane, defined by experts, was considered the 'truth plane'. The 'position error' index of mirrored landmarks was created to quantitatively evaluate the difference among the PWPA SRPs and the truth plane, including overall differences and regional differences of the face (upper, middle and lower). The 'angle error' values between the SRPs and the truth plane in the experimental groups were also evaluated in this study. Statistics and measurement analyses were used to comprehensively evaluate the clinical suitability of the PWPA algorithms to construct the SRP. The average angle error values between the PWPA SRPs of the two functions and the truth plane were 1.21 ± 0.65° and 1.18 ± 0.62°, which were smaller than those between the PA SRP and the truth plane. The position error values of mirrored landmarks constructed using the PWPA algorithms for the whole face and for each facial partition were lower than those constructed using the PA algorithm. In conclusion, for patients with mandibular deviation, this novel mathematical algorithm provided a more suitable SRP for their 3D facial model, which achieved a result approaching the true effect of experts.

A decision-making algorithm combining the aspect-based sentiment analysis and intuitionistic fuzzy-VIKOR for online hotel reservation.

In the process of hotel reservation on online traveling platforms, online reviews, as a fundamental source where the actual information of a product can be had access to, have been attached with high importance by customers when they have difficulty making a decision on which hotel to pick. However, with enormous amount of online reviews distributed in diverse online traveling platforms, customers tend to have few patience or time to manually read all these reviews and get the exact information they want. Inspired by the widespread application of aspect-based sentiment analysis in the field of data mining, a bidirectional long short-term memory (Bi-LSTM) and attention mechanism based model to predict multiple attributes of a product from online review texts is proposed. Experimental result shows that such Bi-LSTM with attention mechanism model apparently improves the accuracy of the prediction, compared with single LSTM model. Meanwhile, based on the output of the prediction, we analyze and transfer it into a statistical matrix. With an intuitionistic fuzzy compromise decision-making method VIKOR applied, an overall ranking, according to multiple product attributes can be made, in which way to help customers make decisions. To prove the rationality of the algorithm, online hotel reviews from three stream online travelling platforms are crawled as a case.

A novel method for 3D face symmetry reference plane based on weighted Procrustes analysis algorithm.

BACKGROUND: We aimed to establish a novel method, using the weighted Procrustes analysis (WPA) algorithm, which assigns weight to facial anatomical landmarks, to construct a three-dimensional facial symmetry reference plane (SRP) for mandibular deviation patients. METHODS: Three-dimensional facial SRPs were independently extracted from 15 mandibular deviation patients using both our WPA algorithm and the standard PA algorithm. A reference plane was defined to serve as the ground truth. To determine whether the WPA SRP or the PA SRP was closer to the ground truth, we measured the position error of mirrored landmarks, the facial asymmetry index (FAI) error, and the angle error for the global face and each facial third partition. RESULTS: The average angle error between the WPA SRP and the ground truth was 1.66 ± 0.81°, which was smaller than that between the PA SRP and the ground truth. The position error of the mirrored landmarks constructed using the WPA algorithm in the global face (3.64 ± 1.53 mm) and each facial partition was lower than that constructed using the PA algorithm. The average FAI error of the WPA SRP was - 7.77 ± 17.02 mm, which was smaller than that of the PA SRP. CONCLUSIONS: This novel automatic algorithm, based on weighted anatomic landmarks, can provide a more adaptable SRP than the standard PA algorithm when applied to severe mandibular deviation patients and can better simulate the diagnosis strategies of clinical experts.

An approach for medical event detection in Chinese clinical notes of electronic health records.

BACKGROUND: Medical event detection in narrative clinical notes of electronic health records (EHRs) is a task designed for reading text and extracting information. Most of the previous work of medical event detection treats the task as extracting concepts at word granularity, which omits the overall structural information of the clinical notes. In this work, we treat each clinical note as a sequence of short sentences and propose an end-to-end deep neural network framework. METHODS: We redefined the task as a sequence labelling task at short sentence granularity, and proposed a novel tag system correspondingly. The dataset were derived from a third-level grade-A hospital, consisting of 2000 annotated clinical notes according to our proposed tag system. The proposed end-to-end deep neural network framework consists of a feature extractor and a sequence labeller, and we explored different implementations respectively. We additionally proposed a smoothed Viterbi decoder as sequence labeller without additional parameter training, which can be a good alternative to conditional random field (CRF) when computing resources are limited. RESULTS: Our sequence labelling models were compared to four baselines which treat the task as text classification of short sentences. Experimental results showed that our approach significantly outperforms the baselines. The best result was obtained by using the convolutional neural networks (CNNs) feature extractor and the sequential CRF sequence labeller, achieving an accuracy of 92.6%. Our proposed smoothed Viterbi decoder achieved a comparable accuracy of 90.07% with reduced training parameters, and brought more balanced performance across all categories, which means better generalization ability. CONCLUSIONS: Evaluated on our annotated dataset, the comparison results demonstrated the effectiveness of our approach for medical event detection in Chinese clinical notes of EHRs. The best feature extractor is the CNNs feature extractor, and the best sequence labeller is the sequential CRF decoder. And it was empirically verified that our proposed smoothed Viterbi decoder could bring better generalization ability while achieving comparable performance to the sequential CRF decoder.