The electrocardiogram (ECG) is a widely used medical test, consisting of voltage versus time traces collected from surface recordings over the heart
. Here we hypothesized that a deep neural network ...(DNN) can predict an important future clinical event, 1-year all-cause mortality, from ECG voltage-time traces. By using ECGs collected over a 34-year period in a large regional health system, we trained a DNN with 1,169,662 12-lead resting ECGs obtained from 253,397 patients, in which 99,371 events occurred. The model achieved an area under the curve (AUC) of 0.88 on a held-out test set of 168,914 patients, in which 14,207 events occurred. Even within the large subset of patients (n = 45,285) with ECGs interpreted as 'normal' by a physician, the performance of the model in predicting 1-year mortality remained high (AUC = 0.85). A blinded survey of cardiologists demonstrated that many of the discriminating features of these normal ECGs were not apparent to expert reviewers. Finally, a Cox proportional-hazard model revealed a hazard ratio of 9.5 (P < 0.005) for the two predicted groups (dead versus alive 1 year after ECG) over a 25-year follow-up period. These results show that deep learning can add substantial prognostic information to the interpretation of 12-lead resting ECGs, even in cases that are interpreted as normal by physicians.
Atrial fibrillation (AF) is associated with substantial morbidity, especially when it goes undetected. If new-onset AF could be predicted, targeted screening could be used to find it early. We ...hypothesized that a deep neural network could predict new-onset AF from the resting 12-lead ECG and that this prediction may help identify those at risk of AF-related stroke.
We used 1.6 M resting 12-lead digital ECG traces from 430 000 patients collected from 1984 to 2019. Deep neural networks were trained to predict new-onset AF (within 1 year) in patients without a history of AF. Performance was evaluated using areas under the receiver operating characteristic curve and precision-recall curve. We performed an incidence-free survival analysis for a period of 30 years following the ECG stratified by model predictions. To simulate real-world deployment, we trained a separate model using all ECGs before 2010 and evaluated model performance on a test set of ECGs from 2010 through 2014 that were linked to our stroke registry. We identified the patients at risk for AF-related stroke among those predicted to be high risk for AF by the model at different prediction thresholds.
The area under the receiver operating characteristic curve and area under the precision-recall curve were 0.85 and 0.22, respectively, for predicting new-onset AF within 1 year of an ECG. The hazard ratio for the predicted high- versus low-risk groups over a 30-year span was 7.2 (95% CI, 6.9-7.6). In a simulated deployment scenario, the model predicted new-onset AF at 1 year with a sensitivity of 69% and specificity of 81%. The number needed to screen to find 1 new case of AF was 9. This model predicted patients at high risk for new-onset AF in 62% of all patients who experienced an AF-related stroke within 3 years of the index ECG.
Deep learning can predict new-onset AF from the 12-lead ECG in patients with no previous history of AF. This prediction may help identify patients at risk for AF-related strokes.
South East Asia has the highest rate of lowland forest loss of any tropical region, with logging and deforestation for conversion to plantation agriculture being flagged as the most urgent threats. ...Detecting and mapping logging impacts on forest structure is a primary conservation concern, as these impacts feed through to changes in biodiversity and ecosystem functions. Here, we test whether high-spatial resolution satellite remote sensing can be used to map the responses of aboveground live tree biomass (AGB), canopy leaf area index (LAI) and fractional vegetation cover (FCover) to selective logging and deforestation in Malaysian Borneo. We measured these attributes in permanent vegetation plots in rainforest and oil palm plantations across the degradation landscape of the Stability of Altered Forest Ecosystems project. We found significant mathematical relationships between field-measured structure and satellite-derived spectral and texture information, explaining up to 62% of variation in biophysical structure across forest and oil palm plots. These relationships held at different aggregation levels from plots to forest disturbance types and oil palms allowing us to map aboveground biomass and canopy structure across the degradation landscape. The maps reveal considerable spatial variation in the impacts of previous logging, a pattern that was less clear when considering field data alone. Up-scaled maps revealed a pronounced decline in aboveground live tree biomass with increasing disturbance, impacts which are also clearly visible in the field data even a decade after logging. Field data demonstrate a rapid recovery in forest canopy structure with the canopy recovering to pre-disturbance levels a decade after logging. Yet, up-scaled maps show that both LAI and FCover are still reduced in logged compared to primary forest stands and markedly lower in oil palm stands. While uncertainties remain, these maps can now be utilised to identify conservation win–wins, especially when combining them with ongoing biodiversity surveys and measurements of carbon sequestration, hydrological cycles and microclimate.
•Map degradation of forests in Borneo based on biophysical attributes•Linking RapidEye-derived reflectance and texture data to field measured structure•Beta-logistic regression models explained up to 62% of structure variation.•Up-scaled maps reveal high spatial variation in the impacts of previous logging.•Maps will be utilised to identify high-conservation-value forests.
Experts are increasingly concerned by issues regarding the activity level of DNA stains. A case from our burglary-related casework pointed out the need for experiments regarding the persistence of ...DNA when more than one person touched a tool handle. We performed short tandem repeat (STR) analyses for three groups of tools: (1) personal and mock owned tools; (2) tools, which were first “owned” by a first user and then handled in a burglary action by a second user; and (3) tools, which were first owned by a first user and then handled in a moderate action. At least three types of tool handles were included in each of the groups. Every second user handled the tool with and without gloves. In total, 234 samples were analyzed regarding profile completeness of first and second user as well as properties like detectable major profile or mixture attributes. When second users simulated a burglary by using a tool bare handed, we could not detect the first user as major component on their handles but attribute him to the stain in 1/40 cases. When second users broke up the burglary setup using gloves, the first user matched the DNA handle profile in 37% of the cases. Moderate use of mock borrowed tools demonstrated a material-dependent persistence. In total, we observed that the outcome depends mainly on the nature of contact, the handle material, and the user-specific characteristics. This study intends to supplement present knowledge about persistence of touch DNA with a special emphasis on burglary-related cases with two consecutive users and to act as experimental data for an evaluation of the relevance of alleged hypotheses, when such is needed in a court hearing.
Naturally heat-resistant coral populations hold significant potential for facilitating coral reef survival under rapid climate change. However, it remains poorly understood whether they can ...acclimatize to ocean warming when superimposed on their already thermally-extreme habitats. Furthermore, it is unknown whether they can maintain their heat tolerance upon larval dispersal or translocation to cooler reefs. We test this in a long-term mesocosm experiment using stress-resistant corals from thermally-extreme reefs in NW Australia. We show that these corals have a remarkable ability to maintain their heat tolerance and health despite acclimation to 3-6 °C cooler, more stable temperatures over 9 months. However, they are unable to increase their bleaching thresholds after 6-months acclimation to + 1 °C warming. This apparent rigidity in the thermal thresholds of even stress-resistant corals highlights the increasing vulnerability of corals to ocean warming, but provides a rationale for human-assisted migration to restore cooler, degraded reefs with corals from thermally-extreme reefs.
Translocations involving the mixed lineage leukemia-1 are recurrent events in acute leukemia and associate with lymphoid (ALL), myeloid (AML) or mixed lineage (MLL) subtypes. Despite an association ...with ALL in humans, murine MLL fusion models are persistently restricted to AML. We here explored this issue using an inducible mixed lineage leukemia-eleven nineteen leukemia (MLL-ENL) mouse model. Although multiple progenitor cell types with myeloid potential are potent AML leukemia-initiating cells, also the earliest lymphoid progenitors were capable of initiating AML. This ability to evoke a latent myeloid potential in the earliest lymphoid progenitors was lost upon further lymphoid commitment. At the same time, more downstream/committed lymphoid precursors also failed to initiate lymphoid leukemia. Co-expression of MLL-ENL with a constitutively active RAS allele, the most common co-mutation in MLL fusion leukemias, could influence on both disease latency and lineage assignment of developing leukemia in what appears to be a mutation-order-dependent manner. Finally, CEBPB-mediated transdifferentation of committed and otherwise leukemia-incompetent B-cell progenitors imbued these cells with leukemic competence for AML. Therefore, apart from providing detailed insight into the differential responsiveness of candidate target cells to a first-hit MLL fusion event, our data warrants caution to therapeutic approaches based on the concept of transdifferentiation.
Summary
In 242 community-dwelling seniors, supplementation with either 1000 mg of calcium or 1000 mg of calcium plus vitamin D resulted in a decrease in the number of subjects with first falls of 27% ...at month 12 and 39% at month 20. Additionally, parameters of muscle function improved significantly.
Introduction
The efficacy of vitamin D and calcium supplementation on risk of falling in the elderly is discussed controversially. Randomized controlled trials using falls as primary outcome are needed. We investigated long-term effects of calcium and vitamin D on falls and parameters of muscle function in community-dwelling elderly women and men.
Methods
Our study population consisted of 242 individuals recruited by advertisements and mailing lists (mean ± SD age, 77 ± 4 years). All serum 25-hydroxyvitamin D (25OHD) levels were below 78 nmol/l. Individuals received in a double blinded fashion either 1000 mg of calcium or 1000 mg of calcium plus 800 IU of vitamin D per day over a treatment period of 12 months, which was followed by a treatment-free but still blinded observation period of 8 months. Falls were documented using diaries. The study took place in Bad Pyrmont, Germany (latitude 52°) and Graz, Austria (latitude 46°).
Results
Compared to calcium mono, supplementation with calcium plus vitamin D resulted in a significant decrease in the number of subjects with first falls of 27% at month 12 (RR = 0.73; CI = 0.54–0.96) and 39% at month 20 (RR = 0.61; CI = 0.34–0.76). Concerning secondary endpoints, we observed significant improvements in quadriceps strength of 8%, a decrease in body sway of 28%, and a decrease in time needed to perform the TUG test of 11%.
Discussion
Combined calcium and vitamin D supplementation proved superior to calcium alone in reducing the number of falls and improving muscle function in community-dwelling older individuals.