Objective To automatically segment diabetic retinal exudation features from deep learning color fundus images. Methods An applied study. The method of this study is based on the U-shaped network model of the Indian Diabetic Retinopathy Image Dataset (IDRID) dataset, introduces deep residual convolution into the encoding and decoding stages, which can effectively extract seepage depth features, solve overfitting and feature interference problems, and improve the model's feature expression ability and lightweight performance. In addition, by introducing an improved context extraction module, the model can capture a wider range of feature information, enhance the perception ability of retinal lesions, and perform excellently in capturing small details and blurred edges. Finally, the introduction of convolutional triple attention mechanism allows the model to automatically learn feature weights, focus on important features, and extract useful information from multiple scales. Accuracy, recall, Dice coefficient, accuracy and sensitivity were used to evaluate the ability of the model to detect and segment the automatic retinal exudation features of diabetic patients in color fundus images. Results After applying this method, the accuracy, recall, dice coefficient, accuracy and sensitivity of the improved model on the IDRID dataset reached 81.56%, 99.54%, 69.32%, 65.36% and 78.33%, respectively. Compared with the original model, the accuracy and Dice index of the improved model are increased by 2.35% , 3.35% respectively. Conclusion The segmentation method based on U-shaped network can automatically detect and segment the retinal exudation features of fundus images of diabetic patients, which is of great significance for assisting doctors to diagnose diseases more accurately.
Recent studies have introduced attention models for medical visual question answering (MVQA). In medical research, not only is the modeling of “visual attention” crucial, but the modeling of “question attention” is equally significant. To facilitate bidirectional reasoning in the attention processes involving medical images and questions, a new MVQA architecture, named MCAN, has been proposed. This architecture incorporated a cross-modal co-attention network, FCAF, which identifies key words in questions and principal parts in images. Through a meta-learning channel attention module (MLCA), weights were adaptively assigned to each word and region, reflecting the model’s focus on specific words and regions during reasoning. Additionally, this study specially designed and developed a medical domain-specific word embedding model, Med-GloVe, to further enhance the model’s accuracy and practical value. Experimental results indicated that MCAN proposed in this study improved the accuracy by 7.7% on free-form questions in the Path-VQA dataset, and by 4.4% on closed-form questions in the VQA-RAD dataset, which effectively improves the accuracy of the medical vision question answer.
In order to solve the saturation distortion phenomenon of R component in fingertip video image, this paper proposes an iterative threshold segmentation algorithm, which adaptively generates the region to be detected for the R component, and extracts the human pulse signal by calculating the gray mean value of the region to be detected. The original pulse signal has baseline drift and high frequency noise. Combining with the characteristics of pulse signal, a zero phase digital filter is designed to filter out noise interference. Fingertip video images are collected on different smartphones, and the region to be detected is extracted by the algorithm proposed in this paper. Considering that the fingertip’s pressure will be different during each measurement, this paper makes a comparative analysis of pulse signals extracted under different pressures. In order to verify the accuracy of the algorithm proposed in this paper in heart rate detection, a comparative experiment of heart rate detection was conducted. The results show that the algorithm proposed in this paper can accurately extract human heart rate information and has certain portability, which provides certain theoretical help for further development of physiological monitoring application on smartphone platform.
In order to overcome the shortcomings of high false positive rate and poor generalization in the detection of microcalcification clusters regions, this paper proposes a method combining discriminative deep belief networks (DDBNs) to automatically and quickly locate the regions of microcalcification clusters in mammograms. Firstly, the breast region was extracted and enhanced, and the enhanced breast region was segmented to overlapped sub-blocks. Then the sub-block was subjected to wavelet filtering. After that, DDBNs model for breast sub-block feature extraction and classification was constructed, and the pre-trained DDBNs was converted to deep neural networks (DNN) using a softmax classifier, and the network is fine-tuned by back propagation. Finally, the undetected mammogram was inputted to complete the location of suspicious lesions. By experimentally verifying 105 mammograms with microcalcifications from the Digital Database for Screening Mammography (DDSM), the method obtained a true positive rate of 99.45% and a false positive rate of 1.89%, and it only took about 16 s to detect a 2 888 × 4 680 image. The experimental results showed that the algorithm of this paper effectively reduced the false positive rate while ensuring a high positive rate. The detection of calcification clusters was highly consistent with expert marks, which provides a new research idea for the automatic detection of microcalcification clusters area in mammograms.
The medical literature contains a wealth of valuable medical knowledge. At present, the research on extraction of entity relationship in medical literature has made great progress, but with the exponential increase in the number of medical literature, the annotation of medical text has become a big problem. In order to solve the problem of manual annotation time such as consuming and heavy workload, a remote monitoring annotation method is proposed, but this method will introduce a lot of noise. In this paper, a novel neural network structure based on convolutional neural network is proposed, which can solve a large number of noise problems. The model can use the multi-window convolutional neural network to automatically extract sentence features. After the sentence vectors are obtained, the sentences that are effective to the real relationship are selected through the attention mechanism. In particular, an entity type (ET) embedding method is proposed for relationship classification by adding entity type characteristics. The attention mechanism at sentence level is proposed for relation extraction in allusion to the unavoidable labeling errors in training texts. We conducted an experiment using 968 medical references on diabetes, and the results showed that compared with the baseline model, the present model achieved good results in the medical literature, and F1-score reached 93.15%. Finally, the extracted 11 types of relationships were stored as triples, and these triples were used to create a medical map of complex relationships with 33 347 nodes and 43 686 relationship edges. Experimental results show that the algorithm used in this paper is superior to the optimal reference system for relationship extraction.
Objective To systematically review the effectiveness and safety of power chain vs. nickel titanium coil springs in closing dental extraction space. Methods Databases including PubMed, EMbase, The Cochrane Library, Chinese Biomedicine Literature Database, Chinese Scientific Journals Full-text Database, and Chinese Journal Full-text Database were searched to collect the randomized controlled trials (RCTs) on comparing power chain with nickel titanium coil springs published before February 2012. Two reviewers independently screened literature, extracted data and assessed the quality of the included studies. Then meta-analysis was conducted using RevMan 5.0 software. Results A total of 4 RCTs involving 122 patients were included. The results of meta-analyses showed that there was a significant difference in the rate of space closure between the two groups (MD=0.30 mm per month, 95%CI 0.17 to 0.44, Plt;0.000 1); The results of subgroup analyses indicated that, both high-quality trials (MD=0.20, 95%CI 0.07 to 0.34, P=0.003) and low quality trials (MD=0.40, 95%CI 0.30 to 0.50, Plt;0.000 01) showed no significant difference in the rate of space closure. Conclusion Current clinical evidence indicates nickel titanium coil spring is superior to power chain in the rate of space closure, but its long-term effect still needs to be proved by more large-scale RCTs.
The chemical extraction method was used to prepare the rat uterine decellularized scaffolds, and to investigate the feasibility of preparing the extracellular matrix (ECM) hydrogel. The rat uterus were collected and extracted by 1%sodium dodecyl sulfate (SDS), 3% TritonX-100 and 4% sodium deoxycholate (SDC) in sequence. Scanning electron microscopy, histochemical staining and immunohistochemistry was used to assess the degree of decellularization of rat uterine scaffold. The prepared decellularized scaffold was digested with pepsin to obtain a uterine ECM hydrogel, and the protein content of ECM was determined by specific ELISA kit. Meanwhile, the mechanical characteristic of ECM hydrogel was measured. The results showed that the chemical extraction method can effectively remove the cells effectively in the rat uterine decellularized scaffold, with the ECM composition preserved completely. ECM hydrogel contains a large amount of ECM protein and shows a good stability, which provides a suitable supporting material for the reconstruction of endometrium in vitro.
Feature extraction is a very crucial step in P300-based brain-computer interface (BCI) and independent component analysis (ICA) is a suitable P300 feature extraction method. But at present the convergence performance of the general ICA iteration methods are not very satisfactory. In this paper, a method based on quantum particle swarm optimizer (QPSO) algorithm and ICA technique is put forward for P300 extraction. In this method, quantum computing is used to impel ICA iteration to globally converge faster. It achieved the purpose of extracting P300 rapidly and efficiently. The method was tested on two public datasets of BCI Competition Ⅱ and Ⅲ, and a simple linear classifier was employed to classify the extracted P300 features. The recognition accuracy reached 94.4% with 15 times averaged. The results showed that the proposed method could extract P300 rapidly and the extraction effect did not reduce. It provides an experimental basis for further study of real-time BCI system.
The purpose of using brain-computer interface (BCI) is to build a bridge between brain and computer for the disable persons, in order to help them to communicate with the outside world. Electroencephalography (EEG) has low signal to noise ratio (SNR), and there exist some problems in the traditional methods for the feature extraction of EEG, such as low classification accuracy, lack of spatial information and huge amounts of features. To solve these problems, we proposed a new method based on time domain, frequency domain and space domain. In this study, independent component analysis (ICA) and wavelet transform were used to extract the temporal, spectral and spatial features from the original EEG signals, and then the extracted features were classified with the method combined support vector machine (SVM) with genetic algorithm (GA). The proposed method displayed a better classification performance, and made the mean accuracy of the Graz datasets in the BCI Competitions of 2003 reach 96%. The classification results showed that the proposed method with the three domains could effectively overcome the drawbacks of the traditional methods based solely on time-frequency domain when the EEG signals were used to describe the characteristics of the brain electrical signals.
Myocardial infarction (MI) has the characteristics of high mortality rate, strong suddenness and invisibility. There are problems such as the delayed diagnosis, misdiagnosis and missed diagnosis in clinical practice. Electrocardiogram (ECG) examination is the simplest and fastest way to diagnose MI. The research on MI intelligent auxiliary diagnosis based on ECG is of great significance. On the basis of the pathophysiological mechanism of MI and characteristic changes in ECG, feature point extraction and morphology recognition of ECG, along with intelligent auxiliary diagnosis method of MI based on machine learning and deep learning are all summarized. The models, datasets, the number of ECG, the number of leads, input modes, evaluation methods and effects of different methods are compared. Finally, future research directions and development trends are pointed out, including data enhancement of MI, feature points and dynamic features extraction of ECG, the generalization and clinical interpretability of models, which are expected to provide references for researchers in related fields of MI intelligent auxiliary diagnosis.