Clinical knowledge-guided deep reinforcement learning for sepsis antibiotic dosing recommendations.

Sepsis is the third leading cause of death worldwide. Antibiotics are an important component in the treatment of sepsis. The use of antibiotics is currently facing the challenge of increasing antibiotic resistance (Evans et al., 2021). Sepsis medication prediction can be modeled as a Markov decision process, but existing methods fail to integrate with medical knowledge, making the decision process potentially deviate from medical common sense and leading to underperformance. (Wang et al., 2021). In this paper, we use Deep Q-Network (DQN) to construct a Sepsis Anti-infection DQN (SAI-DQN) model to address the challenge of determining the optimal combination and duration of antibiotics in sepsis treatment. By setting sepsis clinical knowledge as reward functions to guide DQN complying with medical guidelines, we formed personalized treatment recommendations for antibiotic combinations. The results showed that our model had a higher average value for decision-making than clinical decisions. For the test set of patients, our model predicts that 79.07% of patients will achieve a favorable prognosis with the recommended combination of antibiotics. By statistically analyzing decision trajectories and drug action selection, our model was able to provide reasonable medication recommendations that comply with clinical practices. Our model was able to improve patient outcomes by recommending appropriate antibiotic combinations in line with certain clinical knowledge.

Embracing limited and imperfect training datasets: opportunities and challenges in plant disease recognition using deep learning.

Recent advancements in deep learning have brought significant improvements to plant disease recognition. However, achieving satisfactory performance often requires high-quality training datasets, which are challenging and expensive to collect. Consequently, the practical application of current deep learning-based methods in real-world scenarios is hindered by the scarcity of high-quality datasets. In this paper, we argue that embracing poor datasets is viable and aims to explicitly define the challenges associated with using these datasets. To delve into this topic, we analyze the characteristics of high-quality datasets, namely, large-scale images and desired annotation, and contrast them with the limited and imperfect nature of poor datasets. Challenges arise when the training datasets deviate from these characteristics. To provide a comprehensive understanding, we propose a novel and informative taxonomy that categorizes these challenges. Furthermore, we offer a brief overview of existing studies and approaches that address these challenges. We point out that our paper sheds light on the importance of embracing poor datasets, enhances the understanding of the associated challenges, and contributes to the ambitious objective of deploying deep learning in real-world applications. To facilitate the progress, we finally describe several outstanding questions and point out potential future directions. Although our primary focus is on plant disease recognition, we emphasize that the principles of embracing and analyzing poor datasets are applicable to a wider range of domains, including agriculture. Our project is public available at https://github.com/xml94/EmbracingLimitedImperfectTrainingDatasets.

Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition.

Deep learning shows its advantages and potentials in plant disease recognition and has witnessed a profound development in recent years. To obtain a competing performance with a deep learning algorithm, enough amount of annotated data is requested but in the natural world, scarce or imbalanced data are common, and annotated data is expensive or hard to collect. Data augmentation, aiming to create variations for training data, has shown its power for this issue. But there are still two challenges: creating more desirable variations for scarce and imbalanced data, and designing a data augmentation to ease object detection and instance segmentation. First, current algorithms made variations only inside one specific class, but more desirable variations can further promote performance. To address this issue, we propose a novel data augmentation paradigm that can adapt variations from one class to another. In the novel paradigm, an image in the source domain is translated into the target domain, while the variations unrelated to the domain are maintained. For example, an image with a healthy tomato leaf is translated into a powdery mildew image but the variations of the healthy leaf are maintained and transferred into the powdery mildew class, such as types of tomato leaf, sizes, and viewpoints. Second, current data augmentation is suitable to promote the image classification model but may not be appropriate to alleviate object detection and instance segmentation model, mainly because the necessary annotations can not be obtained. In this study, we leverage a prior mask as input to tell the area we are interested in and reuse the original annotations. In this way, our proposed algorithm can be utilized to do the three tasks simultaneously. Further, We collect 1,258 images of tomato leaves with 1,429 instance segmentation annotations as there is more than one instance in one single image, including five diseases and healthy leaves. Extensive experimental results on the collected images validate that our new data augmentation algorithm makes useful variations and contributes to improving performance for diverse deep learning-based methods.

User Perceptions of Virtual Hospital Apps in China: Systematic Search.

BACKGROUND: Virtual hospital apps are mobile apps that offer functionalities of online consultation, medical guidance, health community forums, referrals, outpatient appointments or virtual hospital-to-home care services. With an increasing number of online medical and health care consulting services, virtual hospital apps have made health care more accessible and fairer for all, especially in China. However, they have occurred without control or regulation. User evaluation can provide directions to help apps optimize identification, lower risks, and guarantee service quality. OBJECTIVE: We aimed to conduct a systematic search for virtual hospital apps in China. To get a global view, virtual hospital apps were assessed and characterized by means of quantitative analysis. To get a local view, we conducted a content feedback analysis to explore user requirements, expectations, and preferences. METHODS: A search was conducted of the most popular Apple and Android app stores in China. We characterized and verified virtual hospital apps and grouped apps according to quantification analysis. We then crawled apps and paid attention to corresponding reviews to incorporate users' involvement, and then performed aspect-based content labeling and analysis using an inductive approach. RESULTS: A total of 239 apps were identified in the virtual hospital app markets in China, and 2686 informative corresponding reviews were analyzed. The evidence showed that usefulness and ease of use were vital facts for engagement. Users were likely to trust a consulting service with a high number of downloads. Furthermore, users expected frequently used apps with more optimization to improve virtual service. We characterized apps according to 4 key features: (1) app functionalities, including online doctor consultation, in-app purchases, tailored education, and community forums; (2) security and privacy, including user data management and user privacy; (3) health management, including health tracking, reminders, and notifications; and (4) technical aspects, including user interface and equipment connection. CONCLUSIONS: Virtual hospitals relying on the mobile internet are growing rapidly. A large number of virtual hospital apps are available and accessible to a growing number of people. Evidence from this systematic search can help various types of virtual hospital models enhance virtual health care experiences, go beyond offline hospitals, and continuously meet the needs of individual end users.

Improved Real-Time Facial Expression Recognition Based on a Novel Balanced and Symmetric Local Gradient Coding.

In the field of Facial Expression Recognition (FER), traditional local texture coding methods have a low computational complexity, while providing a robust solution with respect to occlusion, illumination, and other factors. However, there is still need for improving the accuracy of these methods while maintaining their real-time nature and low computational complexity. In this paper, we propose a feature-based FER system with a novel local texture coding operator, named central symmetric local gradient coding (CS-LGC), to enhance the performance of real-time systems. It uses four different directional gradients on 5 × 5 grids, and the gradient is computed in the center-symmetric way. The averages of the gradients are used to reduce the sensitivity to noise. These characteristics lead to symmetric of features by the CS-LGC operator, thus providing a better generalization capability in comparison to existing local gradient coding (LGC) variants. The proposed system further transforms the extracted features into an eigen-space using a principal component analysis (PCA) for better representation and less computation; it estimates the intended classes by training an extreme learning machine. The recognition rate for the JAFFE database is 95.24%, whereas that for the CK+ database is 98.33%. The results show that the system has advantages over the existing local texture coding methods.

Intensity Variation Normalization for Finger Vein Recognition Using Guided Filter Based Singe Scale Retinex.

Finger vein recognition has been considered one of the most promising biometrics for personal authentication. However, the capacities and percentages of finger tissues (e.g., bone, muscle, ligament, water, fat, etc.) vary person by person. This usually causes poor quality of finger vein images, therefore degrading the performance of finger vein recognition systems (FVRSs). In this paper, the intrinsic factors of finger tissue causing poor quality of finger vein images are analyzed, and an intensity variation (IV) normalization method using guided filter based single scale retinex (GFSSR) is proposed for finger vein image enhancement. The experimental results on two public datasets demonstrate the effectiveness of the proposed method in enhancing the image quality and finger vein recognition accuracy.

Efficient descriptor of histogram of salient edge orientation map for finger vein recognition.

Finger vein images are rich in orientation and edge features. Inspired by the edge histogram descriptor proposed in MPEG-7, this paper presents an efficient orientation-based local descriptor, named histogram of salient edge orientation map (HSEOM). HSEOM is based on the fact that human vision is sensitive to edge features for image perception. For a given image, HSEOM first finds oriented edge maps according to predefined orientations using a well-known edge operator and obtains a salient edge orientation map by choosing an orientation with the maximum edge magnitude for each pixel. Then, subhistograms of the salient edge orientation map are generated from the nonoverlapping submaps and concatenated to build the final HSEOM. In the experiment of this paper, eight oriented edge maps were used to generate a salient edge orientation map for HSEOM construction. Experimental results on our available finger vein image database, MMCBNU_6000, show that the performance of HSEOM outperforms that of state-of-the-art orientation-based methods (e.g., Gabor filter, histogram of oriented gradients, and local directional code). Furthermore, the proposed HSEOM has advantages of low feature dimensionality and fast implementation for a real-time finger vein recognition system.

Robust finger vein ROI localization based on flexible segmentation.

Finger veins have been proved to be an effective biometric for personal identification in the recent years. However, finger vein images are easily affected by influences such as image translation, orientation, scale, scattering, finger structure, complicated background, uneven illumination, and collection posture. All these factors may contribute to inaccurate region of interest (ROI) definition, and so degrade the performance of finger vein identification system. To improve this problem, in this paper, we propose a finger vein ROI localization method that has high effectiveness and robustness against the above factors. The proposed method consists of a set of steps to localize ROIs accurately, namely segmentation, orientation correction, and ROI detection. Accurate finger region segmentation and correct calculated orientation can support each other to produce higher accuracy in localizing ROIs. Extensive experiments have been performed on the finger vein image database, MMCBNU_6000, to verify the robustness of the proposed method. The proposed method shows the segmentation accuracy of 100%. Furthermore, the average processing time of the proposed method is 22 ms for an acquired image, which satisfies the criterion of a real-time finger vein identification system.

Fusion of CT and MR images using an improved wavelet based method.

The fusion of multimodal medical images plays an important role in many clinical applications as it can support more accurate information than any individual source image. This paper presents a novel approach for fusion of computed tomography (CT) and magnetic resonance (MR) images based on wavelet transform. The medical images to be fused are firstly decomposed into multiscale representations by the wavelet transform. Then, by considering the physical meaning of wavelet coefficients and the characteristics of the CT and MR images, the coefficients of the low frequency band and high frequency bands are treated with different schemes: the former is performed with a maximum-selection (MS) rule, and the latter is convolved with a Laplacian operator followed by a MS rule. Finally, the fused image is reconstructed by using the inverse wavelet transform with the combined wavelet coefficients. The performance of our method is qualitatively and quantitatively compared with some existing fusion approaches. The experimental results can demonstrate that the proposed method is a promising and effective technique for fusion of CT and MR images.