Stabilizing Zn/electrolyte Interphasial Chemistry by a Sustained-Release Drug Inspired Indium-Chelated Resin Protective Layer for High-Areal-Capacity Zn//V2O5 Batteries.

For zinc-metal batteries, the instable chemistry at Zn/electrolyte interphasial region results in severe hydrogen evolution reaction (HER) and dendrite growth, significantly impairing Zn anode reversibility. Moreover, an often-overlooked aspect is this instability can be further exacerbated by the interaction with dissolved cathode species in full batteries. Here, inspired by sustained-release drug technology, an indium-chelated resin protective layer (Chelex-In), incorporating a sustained-release mechanism for indium, is developed on Zn surface, stabilizing the anode/electrolyte interphase to ensure reversible Zn plating/stripping performance throughout the entire lifespan of Zn//V2O5 batteries. The sustained-release indium onto Zn electrode promotes a persistent anticatalytic effect against HER and fosters uniform heterogeneous Zn nucleation. Meanwhile, on the electrolyte side, the residual resin matrix with immobilized iminodiacetates anions can also repel detrimental anions (SO4 2- and polyoxovanadate ions dissolved from V2O5 cathode) outside the electric double layer. This dual synergetic regulation on both electrode and electrolyte sides culminates a more stable interphasial environment, effectively enhancing Zn anode reversibility in practical high-areal-capacity full battery systems. Consequently, the bio-inspired Chelex-In protective layer enables an ultralong lifespan of Zn anode over 2800 h, which is also successfully demonstrated in ultrahigh areal capacity Zn//V2O5 full batteries (4.79 mAh cm-2).

Adaptive model training strategy for continuous classification of time series.

The classification of time series is essential in many real-world applications like healthcare. The class of a time series is usually labeled at the final time, but more and more time-sensitive applications require classifying time series continuously. For example, the outcome of a critical patient is only determined at the end, but he should be diagnosed at all times for timely treatment. For this demand, we propose a new concept, Continuous Classification of Time Series (CCTS). Different from the existing single-shot classification, the key of CCTS is to model multiple distributions simultaneously due to the dynamic evolution of time series. But the deep learning model will encounter intertwined problems of catastrophic forgetting and over-fitting when learning multi-distribution. In this work, we found that the well-designed distribution division and replay strategies in the model training process can help to solve the problems. We propose a novel Adaptive model training strategy for CCTS (ACCTS). Its adaptability represents two aspects: (1) Adaptive multi-distribution extraction policy. Instead of the fixed rules and the prior knowledge, ACCTS extracts data distributions adaptive to the time series evolution and the model change; (2) Adaptive importance-based replay policy. Instead of reviewing all old distributions, ACCTS only replays important samples adaptive to their contribution to the model. Experiments on four real-world datasets show that our method outperforms all baselines.

A Series of Polyoxometalate-Viologen Photochromic Materials for UV Probing, Amine Detecting and Inkless and Erasable Printing.

In this work, by using two kinds of viologen ligands three POM-based Compounds were obtained under hydrothermal conditions, namely [AgI (bmypd)0.5 (ß-Mo8 O26 )0.5 ] (1) (bmypd ⋅ 2Cl=1,1'-[Biphenyl-4,4'-bis(methylene)]bis(4,4'-bipyridyinium)dichloride), [AgI 2 (bypy)4 (HSiW12 O40 )2 ] ⋅ 14H2 O (2) and [AgI (bypy)(γ-Mo8 O26 )0.5 ] (3) (bypy⋅Cl=1-Benzyl-4,4'-bipyridyinium chloride). The structures were characterized by Fourier transform infrared spectroscopy (FT-IR), Powder X-ray diffraction (PXRD), X-ray photoelectron spectroscopy (XPS) and single crystal X-ray diffraction. Compounds 1-3 show excellent photochromic ability with fast photoresponse under the irradiation of ultraviolet light with different degrees of color changes. So compounds 1-3 can be used as visible ultraviolet detectors. Compounds 1-3 also possess photoluminescence properties with fast and excellent fluorescence quenching effect. Compounds 1-3 also can be used as inkless and erasable printing materials with suspensions of 1-3 applied to filter paper. Compounds 1-3 can also produce color changes in amine vapor environment, especially in an NH3 atmosphere. Compounds 1-3 can be used as organic amine detectors.

A method for measuring the experimental resolution of laboratory assays (clinical biochemical, blood count, immunological, and qPCR) to evaluate analytical performance.

BACKGROUND: The measurement method for experimental resolution and related data to evaluate analytical performance is poorly explored in clinical research. We established a method to measure the experimental resolution of clinical tests, including biochemical tests, automatic hematology analyzer methods, immunoassays, chemical experiments, and qPCR, to evaluate their analytical performance. METHODS: Serially diluted samples in equal proportions were measured, and correlation analysis was performed between the relative concentration and the measured value. Results were accepted for p ≤ 0.01 of the correlation coefficient. The minimum concentration gradient (eg, 10%) was defined as the experimental resolution. For this method, the smaller the value, the higher the experimental resolution and the better the analytical performance. RESULTS: The experimental resolution of the most common biochemical indices reached 10%, with some even reaching 1%. The results of most counting experiments showed experimental resolution up to 10%, whereas the experimental resolution of the classical chemical assays reached 1%. Unexpectedly, the experimental resolution of more sensitive assays, such as immunoassays was only 25% when using the manual method and 10% for qPCR. CONCLUSION: This study established a method for measuring the experimental resolution of laboratory assays and provides a new index for evaluating the reliability of methods in clinical laboratories.

Interpretable time-aware and co-occurrence-aware network for medical prediction.

BACKGROUND: Disease prediction based on electronic health records (EHRs) is essential for personalized healthcare. But it's hard due to the special data structure and the interpretability requirement of methods. The structure of EHR is hierarchical: each patient has a sequence of admissions, and each admission has some co-occurrence diagnoses. However, the existing methods only partially model these characteristics and lack the interpretation for non-specialists. METHODS: This work proposes a time-aware and co-occurrence-aware deep learning network (TCoN), which is not only suitable for EHR data structure but also interpretable: the co-occurrence-aware self-attention (CS-attention) mechanism and time-aware gated recurrent unit (T-GRU) can model multilevel relations; the interpretation path and the diagnosis graph can make the result interpretable. RESULTS: The method is tested on a real-world dataset for mortality prediction, readmission prediction, disease prediction, and next diagnoses prediction. Experimental results show that TCoN is better than baselines with 2.01% higher accuracy. Meanwhile, the method can give the interpretation of causal relationships and the diagnosis graph of each patient. CONCLUSIONS: This work proposes a novel model-TCoN. It is an interpretable and effective deep learning method, that can model the hierarchical medical structure and predict medical events. The experiments show that it outperforms all state-of-the-art methods. Future work can apply the graph embedding technology based on more knowledge data such as doctor notes.

Predicting COVID-19 disease progression and patient outcomes based on temporal deep learning.

BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic has caused health concerns worldwide since December 2019. From the beginning of infection, patients will progress through different symptom stages, such as fever, dyspnea or even death. Identifying disease progression and predicting patient outcome at an early stage helps target treatment and resource allocation. However, there is no clear COVID-19 stage definition, and few studies have addressed characterizing COVID-19 progression, making the need for this study evident. METHODS: We proposed a temporal deep learning method, based on a time-aware long short-term memory (T-LSTM) neural network and used an online open dataset, including blood samples of 485 patients from Wuhan, China, to train the model. Our method can grasp the dynamic relations in irregularly sampled time series, which is ignored by existing works. Specifically, our method predicted the outcome of COVID-19 patients by considering both the biomarkers and the irregular time intervals. Then, we used the patient representations, extracted from T-LSTM units, to subtype the patient stages and describe the disease progression of COVID-19. RESULTS: Using our method, the accuracy of the outcome of prediction results was more than 90% at 12 days and 98, 95 and 93% at 3, 6, and 9 days, respectively. Most importantly, we found 4 stages of COVID-19 progression with different patient statuses and mortality risks. We ranked 40 biomarkers related to disease and gave the reference values of them for each stage. Top 5 is Lymph, LDH, hs-CRP, Indirect Bilirubin, Creatinine. Besides, we have found 3 complications - myocardial injury, liver function injury and renal function injury. Predicting which of the 4 stages the patient is currently in can help doctors better assess and cure the patient. CONCLUSIONS: To combat the COVID-19 epidemic, this paper aims to help clinicians better assess and treat infected patients, provide relevant researchers with potential disease progression patterns, and enable more effective use of medical resources. Our method predicted patient outcomes with high accuracy and identified a four-stage disease progression. We hope that the obtained results and patterns will aid in fighting the disease.

Multi-Type Sensor Placements in Gaussian Spatial Fields for Environmental Monitoring.

As citizens are increasingly concerned about the surrounding environment, it is important for modern cities to provide sufficient and accurate environmental information to the public for decision making in the era of smart cities. Due to the limited budget, we often need to optimize the sensor placement in order to maximize the overall information gain according to certain criteria. Existing work is primarily concerned with single-type sensor placement; however, the environment usually requires accurate measurements of multiple types of environmental characteristics. In this paper, we focus on the optimal multi-type sensor placement in Gaussian spatial field for environmental monitoring. We study two representative cases: the one-with-all case when each station is equipped with all types of sensors and the general case when each station is equipped with at least one type of sensor. We propose two greedy algorithms accordingly, each with a provable approximation guarantee. We evaluated the proposed approach via an application in air quality monitoring scenario in Hong Kong and experimental results demonstrate the effectiveness of the proposed approach.

miR-32-5p-mediated Dusp5 downregulation contributes to neuropathic pain.

Previous studies have demonstrated that microRNAs (miRNAs) play important roles in the pathogenesis of neuropathic pain. In the present study, we found that miR-32-5p was significantly upregulated in rats after spinal nerve ligation (SNL), specifically in the spinal microglia of rats with SNL. Functional assays showed that knockdown of miR-32-5p greatly suppressed mechanical allodynia and heat hyperalgesia, and decreased inflammatory cytokine (IL-1ß, TNF-α and IL-6) protein expression in rats after SNL. Similarly, miR-32-5p knockdown alleviated cytokine production in lipopolysaccharide (LPS)-treated spinal microglial cells, whereas its overexpression had the opposite effect. Mechanistic investigations revealed Dual-specificity phosphatase 5 (Dusp5) as a direct target of miR-32-5p, which is involved in the miR-32-5p-mediated effects on neuropathic pain and neuroinflammation. We demonstrated for the first time that miR-32-5p promotes neuroinflammation and neuropathic pain development through regulation of Dusp5. Our findings highlight a novel contribution of miR-32-5p to the process of neuropathic pain, and suggest possibilities for the development of novel therapeutic options for neuropathic pain.

Rejuvenation of MPTP-induced human neural precursor cell senescence by activating autophagy.

Aging of neural stem cell, which can affect brain homeostasis, may be caused by many cellular mechanisms. Autophagy dysfunction was found in aged and neurodegenerative brains. However, little is known about the relationship between autophagy and human neural stem cell (hNSC) aging. The present study used 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP) to treat neural precursor cells (NPCs) derived from human embryonic stem cell (hESC) line H9 and investigate related molecular mechanisms involved in this process. MPTP-treated NPCs were found to undergo premature senescence [determined by increased senescence-associated-ß-galactosidase (SA-ß-gal) activity, elevated intracellular reactive oxygen species level, and decreased proliferation] and were associated with impaired autophagy. Additionally, the cellular senescence phenotypes were manifested at the molecular level by a significant increase in p21 and p53 expression, a decrease in SOD2 expression, and a decrease in expression of some key autophagy-related genes such as Atg5, Atg7, Atg12, and Beclin 1. Furthermore, we found that the senescence-like phenotype of MPTP-treated hNPCs was rejuvenated through treatment with a well-known autophagy enhancer rapamycin, which was blocked by suppression of essential autophagy gene Beclin 1. Taken together, these findings reveal the critical role of autophagy in the process of hNSC aging, and this process can be reversed by activating autophagy.

Non-crosslinked systems modulate the gel behavior and structural properties of chitosan/silica composite aerogels.

The aim of this study was to achieve rapid gelation of chitosan (CS) and silica (SA) without crosslinking agent, the relationship between process parameters and the composite aerogels properties were also explored. By varying the composition ratio of the system (from SA:CS = 1:1 to 5:1), the system gelation time was reduced by >12 times, and the drying shrinkage of the composite aerogel reached a minimum of 7.6 %. During the two recombination processes, chitosan rapidly formed aqueous colloid secondary structure under the influence of ethanol. This phenomenon reduced the stability of the system and allowed silica to form a two-phase composite hydrogel. Because the network gap between the fibers was used as a limiting medium for gel growth. In addition, the chitosan/silica composite aerogels exhibited a mesoporous structure with low density (0.1144 g/cm3), and the thermal conductivity was 0.028 W/(m·K) at 30 °C. The trimethylchlorosilane made the composite aerogel have good hydrophobicity with water contact angle as 134.7°, and the adsorption capacity of carbon tetrachloride could reach >10 times of its own weight. This study provides an eco-friendly and high-efficiency method for preparing aerogels, which has potential applications in the fields of thermal insulation, oil-water separation, etc.

Evaluation and analysis of surgical treatment for single-level or multi-level lumbar degenerative disease based on radiography.

Background: Radiography has a low level of radiation exposure while providing valuable information. Due to its cost effectiveness and widespread availability, the preoperative radiographic imaging examination is a valuable approach for assessing patients with spinal disease. This study aimed to examine the influence of preoperative X-ray evaluation on the surgical treatment of patients with single- or multi-level lumbar degenerative disease (LDD). Methods: A retrospective cohort analysis was conducted of 172 patients diagnosed with LDD who underwent transforaminal lumbar interbody fusion (TLIF) or posterior lumbar interbody fusion (PLIF) surgery between December 2021 and February 2023 at the Shanghai Changzheng Hospital. Various parameters were measured on preoperative radiographs, including the iliac crest height, median iliac angle (MIA), lumbar lordosis (LL), intervertebral facet joint degeneration, lumbosacral angle (LSA), intervertebral foramen height (IFH), and surgical segment. The surgical treatment was evaluated based on the operative time, intraoperative blood loss, and postoperative complications. A correlation analysis and independent sample t-tests were used to assess the relationship between preoperative radiographic variables and surgical treatments. Further, a multivariate linear regression analysis was employed to identify the risk factors affecting the clinical outcomes. Results: The correlation analysis and t-test results showed that the MIA, height of the iliac crest, intervertebral facet joint degeneration, and surgical segment were significantly correlated with the surgical treatments (P<0.05). Specifically, the height of the iliac crest, intervertebral facet joint degeneration, and surgical segment were positively correlated with the surgical treatments. Conversely, the MIA was negatively correlated with the surgical treatments. However, no significant differences were observed between the IFH, LSA, and LL in relation to posterior lumbar surgery (P>0.05). The multiple linear regression analysis showed that the height of the iliac crest, MIA, intervertebral facet joint degeneration, and surgical segment were independent factors affecting the surgical treatments of patients with single- or multi-level LDD. These findings highlight the importance of considering these factors when planning and performing lumbar surgery. Conclusions: The measurements taken from radiographs, including the height of the iliac crest, MIA, intervertebral facet joint degeneration, and surgical segment, demonstrate potential influences on the treatment of single- and multi-level lumbar spine surgery. These variables can be captured in plain film imaging and can provide valuable insights into the surgical procedure and offer guidance for the operation. By analyzing these radiographic measurements, surgeons can gain a better understanding of a patient's condition and tailor the surgical approach accordingly, thus optimizing the outcomes of the surgery.

Nitrous Oxide to Reduce Wound Care-Related Pain in Adults: A Systematic Review and Meta-Analysis.

Significance: As an essential procedure, wound care comes with acute pain, which is short but high in intensity, causing patients to fear and affecting subsequent treatment. Nitrous oxide (N2O) is used to relieve pain related to wound care; however, evidence regarding its application is conflicting. Thus, this systematic review and meta-analysis was performed to evaluate the efficacy of N2O in wound care-related pain. Recent Advances: Randomized controlled trials that investigated the effect of N2O in adults undergoing wound care were systematically searched from PubMed, Embase, the Cochrane Library, Web of Science, Scopus, and ClinicalTrials.gov up to February 2023. The primary outcome was the pain score. Secondary outcomes included patients' satisfaction and side effects. Critical Issues: Through screening the 265 identified articles, seven and six studies were finally included in the systematic review and meta-analysis, respectively. Pooled analysis suggested that there was no significant difference in reducing wound care-related pain between the N2O group and the control group (mean difference [MD], -0.02, 95% confidence interval [CI], -1.46, 1.42; p = 0.98, I2 = 96%). Subgroup analyses indicated that there was a significant difference in favor of N2O for burns, not for ulcers, and N2O was superior to oxygen and similar to topical or intravenous anesthesia. There was no significant difference in patients' satisfaction or the incidence of side effects between groups. Future Directions: This review suggests that N2O might be effective for pain management in patients undergoing wound care. Caution must be taken when interpreting these results due to the high risk of biased methods in the included studies.

Facile synthesis of a rod-like Ni/TiO2/C nanocomposite for enhanced electromagnetic wave absorption.

In this study, we utilized a simple calcination method to prepare a Ni/TiO2/C composite, which was synchronously grown from magnetic, semiconductor, and conductive materials. XRD, SEM, Raman, and XPS characterization methods were used to analyze the crystal structure, graphitization degree, morphology size, and valence state of Ni/TiO2/C, and its electromagnetic wave absorption performance was tested. It was revealed that rod-like Ni/TiO2/C had good electromagnetic wave absorption performance at a thickness of 1-5.5 mm; in particular, its reflectance reached -40 dB at 3.5 mm and its absorption bandwidth (reflectivity < -10 dB) reached 4.4 GHz (6.0-10.4 GHz) at a thickness of 4.0 mm. It was thus revealed that its electromagnetic wave absorption rate and absorption bandwidth can be regulated by its thickness. Compared with Ni/TiO2, it was proven that the conductive materials (carbon), magnetic materials (Ni), and semiconductor materials (TiO2) in the rod-like Ni/TiO2/C composite can synergistically absorb electromagnetic wave energy through dielectric and magnetic losses.

[Application and prospects of hyperspectral remote sensing in monitoring plant diversity in grassland]. / é«˜å ‰è°±é¥æ„Ÿåœ¨è‰åŽŸæ¤ç‰©å¤šæ ·æ€§ç›‘æµ‹ä¸­çš„åº”ç”¨ä¸Žå±•æœ›.

The biodiversity of grasslands is important for ecosystem function and health. The protection and mana-gement of grassland biodiversity requires the collection of the information on plant diversity. Hyperspectral remote sensing, with its unique advantages of extensive coverage and high spectral resolution, offers a new solution for long-term monitoring of plant diversity. We first reviewed the development history of hyperspectral remote sensing technology, emphasized its advantages in monitoring grassland plant diversity, and further analyzed its specific applications in this field. Finally, we discussed the challenges faced by hyperspectral remote sensing technology in its applications, such as the complexity of data processing, accuracy of algorithms, and integration with ground-based remote sensing data, and proposes prospects for future research directions. With the advancement of remote sensing technology and the integrated application of multi-source data, hyperspectral remote sensing would play an increasingly important role in grassland ecological monitoring and biodiversity conservation, which could provide scientific basis and technical support for global ecological protection and sustainable development.

Optimizing biomagnetic sensor performance through in silico diagnostics: A novel approach with BEST (Biomagnetism Evaluation via Simulated Testing).

Advancing biomagnetic measurement capabilities requires a nuanced understanding of sensor performance beyond traditional metrics. This study introduces Biomagnetism Evaluation via Simulated Testing (BEST), a novel methodology combining a current dipole model simulating cardiac biomagnetic fields with a convolutional neural network. Our investigation reveals that optimal sensor array performance is achieved when sensors are in close proximity to the magnetic source, with a shorter effective domain. Contrary to common assumptions, the bottom edge length of the sensor has a negligible impact on array performance. BEST provides a versatile framework for exploring the influence of diverse technical indicators on biomagnetic sensor performance, offering valuable insights for sensor development and selection.

Estimating causal effects of physical disability and number of comorbid chronic diseases on risk of depressive symptoms in an elderly Chinese population: a machine learning analysis of cross-sectional baseline data from the China longitudinal ageing social survey.

OBJECTIVE: This study aimed to explore the causal effects of physical disability and number of comorbid chronic diseases on depressive symptoms in an elderly Chinese population. DESIGN, SETTING AND ANALYSIS: Cross-sectional, baseline data were obtained from the China Longitudinal Ageing Social Survey, a stratified, multistage, probabilistic sampling survey conducted in 2014 that covers 28 of 31 provincial areas in China. The causal effects of physical disability and number of comorbid chronic diseases on depressive symptoms were analysed using the conditional average treatment effect method of machine learning. The causal effects model's adjustment was made for age, gender, residence, marital status, educational level, ethnicity, wealth quantile and other factors. OUTCOME: Assessment of the causal effects of physical disability and number of comorbid chronic diseases on depressive symptoms. PARTICIPANTS: 7496 subjects who were 60 years of age or older and who answered the questions on depressive symptoms and other independent variables of interest in a survey conducted in 2014 were included in this study. RESULTS: Physical disability and number of comorbid chronic diseases had causal effects on depressive symptoms. Among the subjects who had one or more functional limitations, the probability of depressive symptoms increased by 22% (95% CI 19% to 24%). For the subjects who had one chronic disease and those who had two or more chronic diseases, the possibility of depressive symptoms increased by 13% (95% CI 10% to 15%) and 20% (95% CI 18% to 22%), respectively. CONCLUSION: This study provides evidence that the presence of one or more functional limitations affects the occurrence of depressive symptoms among elderly people. The findings of our study are of value in developing programmes that are designed to identify elderly individuals who have physical disabilities or comorbid chronic diseases to provide early intervention.

A Ranking-Based Cross-Entropy Loss for Early Classification of Time Series.

Early classification tasks aim to classify time series before observing full data. It is critical in time-sensitive applications such as early sepsis diagnosis in the intensive care unit (ICU). Early diagnosis can provide more opportunities for doctors to rescue lives. However, there are two conflicting goals in the early classification task-accuracy and earliness. Most existing methods try to find a balance between them by weighing one goal against the other. But we argue that a powerful early classifier should always make highly accurate predictions at any moment. The main obstacle is that the key features suitable for classification are not obvious in the early stage, resulting in the excessive overlap of time series distributions in different time stages. The indistinguishable distributions make it difficult for classifiers to recognize. To solve this problem, this article proposes a novel ranking-based cross-entropy () loss to jointly learn the feature of classes and the order of earliness from time series data. In this way, can help classifier to generate probability distributions of time series in different stages with more distinguishable boundary. Thus, the classification accuracy at each time step is finally improved. Besides, for the applicability of the method, we also accelerate the training process by focusing the learning process on high-ranking samples. Experiments on three real-world datasets show that our method can perform classification more accurately than all baselines at all moments.

Safety and efficacy of PD-1 and PD-L1 inhibitors in relapsed and refractory Hodgkin's lymphoma: a systematic review and meta-analysis of 20 prospective studies.

BACKGROUND: Inhibitors of programmed cell death protein 1 (PD-1) and programmed cell death ligand 1 (PD-L1) have been used in the treatment of relapsed and refractory Hodgkin's lymphoma (R/R HL) recently. To further understand the safety and efficacy of PD-1/PD-L1 inhibitors in R/R HL, we conducted this meta-analysis. METHODS: Databases and the Clinical Registration Platforms have been systematically searched for related studies by March 2022. For safety analysis, the incidence and exhibition of any grade and grade 3 or higher adverse effects (AEs) were evaluated. Besides, severe AEs (SAEs), treatment-related deaths, and AEs leading to treatment discontinuation were summarized. The overall response rate (ORR), complete response (CR) rate, partial response (PR) rate, progression-free survival (PFS), overall survival (OS), and duration of response (DOR) were calculated for efficacy analysis. All processes were implemented mainly through the package Meta and MetaSurv of software R 4.1.2. RESULTS: Overall 20 studies and 1440 patients were enrolled. The pooled incidence of any grade and grade 3 or higher AEs were 92% and 26%, respectively. The pooled ORR, CR rate and PR rate were 79%, 44% and 34%, respectively. The most common AEs were neuropathy (29%), nausea (27%), pyrexia (26%), and leukopenia (25%), and the most common grade 3 or higher AEs included leukopenia (10%), infusion reaction (8%), weight gain (3%), and neutropenia (2.7%). In survival analysis, pembrolizumab monotherapy appeared to perform better compared to nivolumab monotherapy. CONCLUSIONS: PD-1/PD-L1 inhibitors show promising efficacy and tolerable AEs in the treatment of R/R HL.

Continuous diagnosis and prognosis by controlling the update process of deep neural networks.

Continuous diagnosis and prognosis are essential for critical patients. They can provide more opportunities for timely treatment and rational allocation. Although deep-learning techniques have demonstrated superiority in many medical tasks, they frequently forget, overfit, and produce results too late when performing continuous diagnosis and prognosis. In this work, we summarize the four requirements; propose a concept, continuous classification of time series (CCTS); and design a training method for deep learning, restricted update strategy (RU). The RU outperforms all baselines and achieves average accuracies of 90%, 97%, and 85% on continuous sepsis prognosis, COVID-19 mortality prediction, and eight disease classifications, respectively. The RU can also endow deep learning with interpretability, exploring disease mechanisms through staging and biomarker discovery. We find four sepsis stages, three COVID-19 stages, and their respective biomarkers. Further, our approach is data and model agnostic. It can be applied to other diseases and even in other fields.