A multi-dimensional fusion strategy similarity measure method for patent application technology disclosure document.

Patent application technology disclosure document is one of the important bases for judging patent novelty and uniqueness. Automated evaluation can effectively solve the problems of long time and strong subjectivity of human evaluation. The text similarity evaluation algorithm based on corpus and deep learning technology has problems such as insufficient amount of cross-library learning data and insufficient core content tendency in the similarity judgment of patent application technology disclosure document, which limits their performance and practical application. In this paper, we propose a similarity evaluation method of patent application technology disclosure document based on multi-dimensional fusion strategy to realize the similarity measurement of patents. Firstly, in the text preprocessing section, word segmentation reconstruction and similarity evaluation optimization strategies based on word frequency and part-of-speech score weighted fusion are proposed. Then, a similarity calculation method of patent application technology disclosure document based on two new mapping spaces of dot matrix and image is proposed to achieve a more diversified comprehensive evaluation. The algorithm was evaluated by using four published text similarity matching datasets (containing 0-5 or 0/1 labels) and a set of patent application technology disclosure documents. Experimental results show that on the published text similarity matching datasets, the similarity evaluation method under the multi-dimensional fusion strategy proposed in this paper has a discrimination accuracy improvement of about 10% compared to traditional vector semantic model, and can match the discriminative ability of lightweight deep learning models without the need for training. At the same time, the discrimination accuracy of the proposed method on the sample dataset of patent application technology disclosure document is superior to traditional vector semantic model (20%) and various deep learning models (1%-8%), and the precision and recall rate are relatively balanced. The visual analysis results on the dataset of the patent application technology disclosure documents also prove the effectiveness and reliability of the similarity calculation method proposed in the dot matrix and image space, which provide a new idea and method for the similarity evaluation between patent application technology disclosure document.

Weakly supervised training for eye fundus lesion segmentation in patients with diabetic retinopathy.

OBJECTIVE: Diabetic retinopathy is the leading cause of vision loss in working-age adults. Early screening and diagnosis can help to facilitate subsequent treatment and prevent vision loss. Deep learning has been applied in various fields of medical identification. However, current deep learning-based lesion segmentation techniques rely on a large amount of pixel-level labeled ground truth data, which limits their performance and application. In this work, we present a weakly supervised deep learning framework for eye fundus lesion segmentation in patients with diabetic retinopathy. METHODS: First, an efficient segmentation algorithm based on grayscale and morphological features is proposed for rapid coarse segmentation of lesions. Then, a deep learning model named Residual-Attention Unet (RAUNet) is proposed for eye fundus lesion segmentation. Finally, a data sample of fundus images with labeled lesions and unlabeled images with coarse segmentation results is jointly used to train RAUNet to broaden the diversity of lesion samples and increase the robustness of the segmentation model. RESULTS: A dataset containing 582 fundus images with labels verified by doctors, including hemorrhage (HE), microaneurysm (MA), hard exudate (EX) and soft exudate (SE), and 903 images without labels was used to evaluate the model. In ablation test, the proposed RAUNet achieved the highest intersection over union (IOU) on the labeled dataset, and the proposed attention and residual modules both improved the IOU of the UNet benchmark. Using both the images labeled by doctors and the proposed coarse segmentation method, the weakly supervised framework based on RAUNet architecture significantly improved the mean segmentation accuracy by over 7% on the lesions. SIGNIFICANCE: This study demonstrates that combining unlabeled medical images with coarse segmentation results can effectively improve the robustness of the lesion segmentation model and proposes a practical framework for improving the performance of medical image segmentation given limited labeled data samples.

Diagnosing Diabetic Retinopathy in OCTA Images Based on Multilevel Information Fusion Using a Deep Learning Framework.

Objective: As an extension of optical coherence tomography (OCT), optical coherence tomographic angiography (OCTA) provides information on the blood flow status at the microlevel and is sensitive to changes in the fundus vessels. However, due to the distinct imaging mechanism of OCTA, existing models, which are primarily used for analyzing fundus images, do not work well on OCTA images. Effectively extracting and analyzing the information in OCTA images remains challenging. To this end, a deep learning framework that fuses multilevel information in OCTA images is proposed in this study. The effectiveness of the proposed model was demonstrated in the task of diabetic retinopathy (DR) classification. Method: First, a U-Net-based segmentation model was proposed to label the boundaries of large retinal vessels and the foveal avascular zone (FAZ) in OCTA images. Then, we designed an isolated concatenated block (ICB) structure to extract and fuse information from the original OCTA images and segmentation results at different fusion levels. Results: The experiments were conducted on 301 OCTA images. Of these images, 244 were labeled by ophthalmologists as normal images, and 57 were labeled as DR images. An accuracy of 93.1% and a mean intersection over union (mIOU) of 77.1% were achieved using the proposed large vessel and FAZ segmentation model. In the ablation experiment with 6-fold validation, the proposed deep learning framework that combines the proposed isolated and concatenated convolution process significantly improved the DR diagnosis accuracy. Moreover, inputting the merged images of the original OCTA images and segmentation results further improved the model performance. Finally, a DR diagnosis accuracy of 88.1% (95%CI ± 3.6%) and an area under the curve (AUC) of 0.92 were achieved using our proposed classification model, which significantly outperforms the state-of-the-art classification models. As a comparison, an accuracy of 83.7 (95%CI ± 1.5%) and AUC of 0.76 were obtained using EfficientNet. Significance. The visualization results show that the FAZ and the vascular region close to the FAZ provide more information for the model than the farther surrounding area. Furthermore, this study demonstrates that a clinically sophisticated designed deep learning model is not only able to effectively assist in the diagnosis but also help to locate new indicators for certain illnesses.