Real-world post-deployment performance of a novel machine learning-based digital health technology for skin lesion assessment and suggestions for post-market surveillance
Ontology highlight
ABSTRACT:
SUBMITTER: Thomas L
PROVIDER: S-EPMC10645139 | biostudies-literature | 2023 Jan
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:With the rapid development of digital technology, digital technology innovation has become a core driver of China's economic development. Thus, this study uses A-share listed companies from 2003 to 2021 as the research sample. The digital patents of firms are identified to portray the level of digital technology innovation by matching the digital economy industry classification code, national economy industry classification code, and IPC number. Considers the economic effect of digital technology innovation from the perspective of firm market value. It is found that digital technology innovation significantly contributes to the increase in firm market value, and this finding still holds when robustness tests are performed. Mechanistic tests have shown that digital technology innovation affects firm market value by driving digital transformation, promoting productivity, and enhancing market profitability. Further analysis reveals that digital technology innovation has a more significant effect on increasing firm market value for large, non-state, capital-intensive, technology-intensive and low internal control costs firms. This study verifies the enabling effect of digital technology innovation on the development of the real economy at the micro level, and provides insights for the optimization of China's digital technology innovation policies and the formulation of firms' digital development strategies.
| S-EPMC11376586 | biostudies-literature
Project description:Stock market prediction is a challenging task as it requires deep insights for extraction of news events, analysis of historic data, and impact of news events on stock price trends. The challenge is further exacerbated due to the high volatility of stock price trends. However, a detailed overview that discusses the overall context of stock prediction is elusive in literature. To address this research gap, this paper presents a detailed survey. All key terms and phases of generic stock prediction methodology along with challenges, are described. A detailed literature review that covers data preprocessing techniques, feature extraction techniques, prediction techniques, and future directions is presented for news sensitive stock prediction. This work investigates the significance of using structured text features rather than unstructured and shallow text features. It also discusses the use of opinion extraction techniques. In addition, it emphasizes the use of domain knowledge with both approaches of textual feature extraction. Furthermore, it highlights the significance of deep neural network based prediction techniques to capture the hidden relationship between textual and numerical data. This survey is significant and novel as it elaborates a comprehensive framework for stock market prediction and highlights the strengths and weaknesses of existing approaches. It presents a wide range of open issues and research directions that are beneficial for the research community.
| S-EPMC8114814 | biostudies-literature
Project description:The use of sequential statistical analysis for post-market drug safety surveillance is quickly emerging. Both continuous and group sequential analysis have been used, but consensus is lacking as to when to use which approach. We compare the statistical performance of continuous and group sequential analysis in terms of type I error probability; statistical power; expected time to signal when the null hypothesis is rejected; and the sample size required to end surveillance without rejecting the null. We present a mathematical proposition to show that for any group sequential design there always exists a continuous sequential design that is uniformly better. As a consequence, it is shown that more frequent testing is always better. Additionally, for a Poisson based probability model and a flat rejection boundary in terms of the log likelihood ratio, we compare the performance of various continuous and group sequential designs. Using exact calculations, we found that, for the parameter settings used, there is always a continuous design with shorter expected time to signal than the best group design. The two key conclusions from this article are (i) that any post-market safety surveillance system should attempt to obtain data as frequently as possible, and (ii) that sequential testing should always be performed when new data arrives without deliberately waiting for additional data.
| S-EPMC4888780 | biostudies-literature
Project description:ObjectiveThe aim of this study was to describe previously unrecognised or under-recognised adverse events associated with Melody® valve implantation.BackgroundIn rare diseases and conditions, it is typically not feasible to conduct large-scale safety trials before drug or device approval. Therefore, post-market surveillance mechanisms are necessary to detect rare but potentially serious adverse events.MethodsWe reviewed the United States Food and Drug Administration's Manufacturer and User Facility Device Experience (MAUDE) database and conducted a structured literature review to evaluate adverse events associated with on- and off-label Melody® valve implantation. Adverse events were compared with those described in the prospective Investigational Device Exemption and Post-Market Approval Melody® transcatheter pulmonary valve trials.ResultsWe identified 631 adverse events associated with "on-label" Melody® valve implants and 84 adverse events associated with "off-label" implants. The most frequent "on-label" adverse events were similar to those described in the prospective trials including stent fracture (n=210) and endocarditis (n=104). Previously unrecognised or under-recognised adverse events included stent fragment embolisation (n=5), device erosion (n=4), immediate post-implant severe valvar insufficiency (n=2), and late coronary compression (n=2 cases at 5 days and 3 months after implantation). Under-recognised adverse events associated with off-label implantation included early valve failure due to insufficiency when implanted in the tricuspid position (n=7) and embolisation with percutaneous implantation in the mitral position (n=5).ConclusionPost-market passive surveillance does not demonstrate a high frequency of previously unrecognised serious adverse events with "on-label" Melody® valve implantation. Further study is needed to evaluate safety of "off-label" uses.
| S-EPMC5425318 | biostudies-literature
Project description:IntroductionInfectious diseases pose a risk to public health, requiring efficient strategies for disease prevention. Digital health surveillance technologies provide new opportunities to enhance disease prevention, detection, tracking, reporting and analysis. However, in addition to concerns regarding the effectiveness of these technologies in meeting public health goals, there are also concerns regarding the ethics, legality, safety and sustainability of digital surveillance technologies. This scoping review examines the literature on digital surveillance for public health purposes during the COVID-19 pandemic to identify health-related applications of digital surveillance technologies, and to highlight discussions of the implications of these technologies.Methods and analysisThe scoping review will be guided by the framework proposed by Arksey and O'Malley and the guidelines outlined by Colquhoun et al and Levac et al. We will search Medline (Ovid), PsycInfo, PubMed, Scopus, CINAHL (EBSCOhost), ACM Digital Library, Google Scholar and IEEE Explore for relevant studies published between December 2019 and December 2020. The review will also include grey literature. Data will be managed and analysed through an extraction table and thematic analysis.Ethics and disseminationFindings will be disseminated through traditional academic channels, as well as social media channels and research briefs and infographics. We will target our dissemination to provincial and federal public health organisations, as well as technology companies and community-based organisations managing the public response to the COVID-19 pandemic.
| S-EPMC8561829 | biostudies-literature
Project description:Trial investigators often have a primary interest in the estimation of the survival curve in a population for which there exists acceptable historical information from which to borrow strength. However, borrowing strength from a historical trial that is non-exchangeable with the current trial can result in biased conclusions. In this article we propose a fully Bayesian semiparametric method for the purpose of attenuating bias and increasing efficiency when jointly modeling time-to-event data from two possibly non-exchangeable sources of information. We illustrate the mechanics of our methods by applying them to a pair of post-market surveillance datasets regarding adverse events in persons on dialysis that had either a bare metal or drug-eluting stent implanted during a cardiac revascularization surgery. We finish with a discussion of the advantages and limitations of this approach to evidence synthesis, as well as directions for future work in this area. The article's Supplementary Materials offer simulations to show our procedure's bias, mean squared error, and coverage probability properties in a variety of settings.
| S-EPMC3954409 | biostudies-literature
Project description:BackgroundValidating new algorithms, such as methods to disentangle intrinsic treatment risk from risk associated with experiential learning of novel treatments, often requires knowing the ground truth for data characteristics under investigation. Since the ground truth is inaccessible in real world data, simulation studies using synthetic datasets that mimic complex clinical environments are essential. We describe and evaluate a generalizable framework for injecting hierarchical learning effects within a robust data generation process that incorporates the magnitude of intrinsic risk and accounts for known critical elements in clinical data relationships.MethodsWe present a multi-step data generating process with customizable options and flexible modules to support a variety of simulation requirements. Synthetic patients with nonlinear and correlated features are assigned to provider and institution case series. The probability of treatment and outcome assignment are associated with patient features based on user definitions. Risk due to experiential learning by providers and/or institutions when novel treatments are introduced is injected at various speeds and magnitudes. To further reflect real-world complexity, users can request missing values and omitted variables. We illustrate an implementation of our method in a case study using MIMIC-III data for reference patient feature distributions.ResultsRealized data characteristics in the simulated data reflected specified values. Apparent deviations in treatment effects and feature distributions, though not statistically significant, were most common in small datasets (n < 3000) and attributable to random noise and variability in estimating realized values in small samples. When learning effects were specified, synthetic datasets exhibited changes in the probability of an adverse outcomes as cases accrued for the treatment group impacted by learning and stable probabilities as cases accrued for the treatment group not affected by learning.ConclusionsOur framework extends clinical data simulation techniques beyond generation of patient features to incorporate hierarchical learning effects. This enables the complex simulation studies required to develop and rigorously test algorithms developed to disentangle treatment safety signals from the effects of experiential learning. By supporting such efforts, this work can help identify training opportunities, avoid unwarranted restriction of access to medical advances, and hasten treatment improvements.
| S-EPMC10088292 | biostudies-literature
Project description:Type I error probability spending functions are commonly used for designing sequential analysis of binomial data in clinical trials, but it is also quickly emerging for near-continuous sequential analysis of post-market drug and vaccine safety surveillance. It is well known that, for clinical trials, when the null hypothesis is not rejected, it is still important to minimize the sample size. Unlike in post-market drug and vaccine safety surveillance, that is not important. In post-market safety surveillance, specially when the surveillance involves identification of potential signals, the meaningful statistical performance measure to be minimized is the expected sample size when the null hypothesis is rejected. The present paper shows that, instead of the convex Type I error spending shape conventionally used in clinical trials, a concave shape is more indicated for post-market drug and vaccine safety surveillance. This is shown for both, continuous and group sequential analysis.
| S-EPMC6482950 | biostudies-literature
Project description:Statistical sequential hypothesis testing is meant to analyze cumulative data accruing in time. The methods can be divided in two types, group and continuous sequential approaches, and a question that arises is if one approach suppresses the other in some sense. For Poisson stochastic processes, we prove that continuous sequential analysis is uniformly better than group sequential under a comprehensive class of statistical performance measures. Hence, optimal solutions are in the class of continuous designs. This paper also offers a pioneer study that compares classical Type I error spending functions in terms of expected number of events to signal. This was done for a number of tuning parameters scenarios. The results indicate that a log-exp shape for the Type I error spending function is the best choice in most of the evaluated scenarios.
| S-EPMC6936745 | biostudies-literature
Project description:ObjectiveCompliance with post-operative follow-up in the context of international surgical trips is often poor. The etiology of this problem is multifactorial and includes lack of local physician involvement, transportation costs, and work responsibilities. We aimed to better understand availability of communication technologies within Belize and use this information to improve follow-up after visiting surgical trips to a public hospital in Belize City. Accordingly, a 6-item questionnaire assessing access to communication technologies was completed by all patients undergoing evaluation by a visiting surgical team in 2014. Based on this data, a pilot program for patients undergoing surgery was instituted for subsequent missions (2015-2016) that included a 6-week post-operative telephone interview with a visiting physician located in the United States.ResultsFifty-four (n = 54) patients were assessed via survey with 89% responding that they had a mobile phone. Patients reported less access to home internet (59%), local internet (52%), and email (48%). Of 35 surgical patients undergoing surgery during 2 subsequent surgical trips, 18 (51%) were compliant with telephone interview at 6-week follow-up. Issues were identified in 3 (17%) patients that allowed for physician assistance. The cost per patient interview was $10 USD.
| S-EPMC5822548 | biostudies-literature