Sensitivity Maps Research Articles

A deep learning algorithm to detect anaemia with ECGs: a retrospective, multicentre study

Anaemia is an important health-care burden globally, and screening for anaemia is crucial to prevent multi-organ injury, irreversible complications, and life-threatening adverse events. We aimed to establish whether a deep learning algorithm (DLA) that enables non-invasive anaemia screening from electrocardiograms (ECGs) might improve the detection of anaemia. We did a retrospective, multicentre, diagnostic study in which a DLA was developed using ECGs and then internally and externally validated. We used data from two hospitals, Sejong General Hospital (hospital A) and Mediplex Sejong Hospital (hospital B), in South Korea. Data from hospital A was for DLA development and internal validation, and data from hospital B was for external validation. We included individuals who had at least one ECG with a haemoglobin measurement within 1 h of the index ECG and excluded individuals with missing demographic, electrocardiographic, or haemoglobin information. Three types of DLA were developed with 12-lead, 6-lead (limb lead), and single-lead (lead I) ECGs to detect haemoglobin concentrations of 10 g/dL or less. The DLA was built by a convolutional neural network and used 500-Hz raw ECG, age, and sex as input data. The study period ran from Oct 1, 2016, to Sept 30, 2019, in hospital A and March 1, 2017, to Sept 30, 2019, in hospital B. 40 513 patients at hospital A and 4737 patients at hospital B were eligible for inclusion. We excluded 281 patients at hospital A and 72 patients at hospital B because of missing values for clinical information and ECG data. The development dataset comprised 57 435 ECGs from 31 898 patients, and the algorithm was internally validated with 7974 ECGs from 7974 patients. The external validation dataset included 4665 ECGs from 4665 patients. 586 (internal) and 194 (external) patients within the combined dataset were found to be anaemic. During internal and external validation, the area under the receiver operating characteristics curve (AUROC) of the DLA using a 12-lead ECG for detecting anaemia was 0·923 for internal validation and 0·901 for external validation. Using a 90% sensitivity operating point for the development data, the sensitivity, specificity, negative predictive value, and positive predictive value of internal validation were 89·8%, 81·5%, 99·4%, and 20·0%, respectively, and those of external validation were 86·1%, 76·2%, 99·2%, and 13·5%, respectively. The DLA focused on the QRS complex for deciding the presence of anaemia in a sensitivity map. The AUROCs of DLAs using 6 leads and a single lead were in the range of 0·841-0·890. In this study, using raw ECG data, a DLA accurately detected anaemia. The application of artificial intelligence to ECGs could enable screening for anaemia. None.

Identifying States of Collateral Sensitivity during the Evolution of Therapeutic Resistance in Ewing's Sarcoma.

SummaryAdvances in the treatment of Ewing's sarcoma (EWS) are desperately needed, particularly in the case of metastatic disease. A deeper understanding of collateral sensitivity, where the evolution of therapeutic resistance to one drug aligns with sensitivity to another drug, may improve our ability to effectively target this disease. For the first time in a solid tumor, we produced a temporal collateral sensitivity map that demonstrates the evolution of collateral sensitivity and resistance in EWS. We found that the evolution of collateral resistance was predictable with some drugs but had significant variation in response to other drugs. Using this map of temporal collateral sensitivity in EWS, we can see that the path toward collateral sensitivity is not always repeatable, nor is there always a clear trajectory toward resistance or sensitivity. Identifying transcriptomic changes that accompany these states of transient collateral sensitivity could improve treatment planning for patients with EWS.

Multi-input deep learning architecture for predicting breast tumor response to chemotherapy using quantitative MR images.

Neoadjuvant chemotherapy (NAC) aims to minimize the tumor size before surgery. Predicting response to NAC could reduce toxicity and delays to effective intervention. Computational analysis of dynamic contrast-enhanced magnetic resonance images (DCE-MRI) through deep convolution neural network (CNN) has shown a significant performance to distinguish responders and no responder's patients. This study intends to present a new deep learning (DL) model predicting the breast cancer response to NAC based on multiple MRI inputs. A cohort of 723 axial slices extracted from 42 breast cancer patients who underwent NAC therapy was used to train and validate the developed DL model. This dataset was provided by our collaborator institute of radiology in Brussels. Fourteen external cases were used to validate the best obtained model to predict pCR based on pre- and post-chemotherapy DCE-MRI. The model performance was assessed by area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and feature map visualization. The developed multi-inputs deep learning architecture was able to predict the pCR to NAC treatment in the validation dataset with an AUC of 0.91 using combined pre- and post-NAC images. The visual results showed that the most important extracted features from non-pCR tumors are in the peripheral region. The proposed method was more productive than the previous ones. Even with a limited training dataset size, the proposed and developed CNN model using DCE-MR images acquired before and after the first chemotherapy was able to classify pCR and non-pCR patients with substantial accuracy. This model could be used hereafter in clinical analysis after its evaluation based on more extra data.

A field study to predict Cd bioaccumulation in a soil-wheat system: Application of a geochemical model

An accurate model to predict Cd accumulation in crops based on soil properties would facilitate evaluations of soil quality and the potential risk posed by metals. However, given the heterogeneity of soil, such models are difficult to establish on large regional scales. This study for the first time examined the applicability of a multi-surface speciation model (MSM) in predicting Cd accumulation in wheat at a regional field scale, based on 140 soil-wheat paired samples collected from a 205-km2 field. The MSM resulted in a better correlation between Cd accumulation in wheat grain (R2 = 0.75) and roots (R2 = 0.74) than obtained with chemical extraction methods (total Cd in soil, 0.01 M CaCl2, and 0.43 M HNO3). In addition, while the performance of the MSM was comparable to that of a traditional multiple regression model, a parameter-fitting process was not required. The predictive ability of the MSM was further used to assess and predict the soil Cd risk and to develop a soil Cd sensitivity map to better localize areas of greatest sensitivity to Cd contamination. The results showed that the MSM can serve as a useful tool for regional soil risk assessments and thus in the development of soil protection measures.

Regional modeling of soil sensitivity to water erosion in JIJEL region (Algeria) using MCA and GIS

Water erosion is a serious problem that threatens more and more soils in northern Algeria and particularly in Jijel region. The potential of its natural and tourist resources makes this region an important ecological and socio-economic area that deserves to be preserved. The main objective of this work is to apply a weighted-sum model to water erosion sensitivity mapping in Jijel region by integrating geographical information system (GIS). Four layers of data modeling factors responsible for soil erosion vulnerability were combined: precipitation, topography, lithology, and vegetation cover. The resulting water-erosion sensitivity map enabled distinguishing areas most threatened by erosion; 77% of the total area was found to be highly susceptible to water erosion, mainly caused by steep slopes and vegetation cover degraded by summer fires. The sensitivity map was validated by the high statistically significant value of chi-square test and is, therefore, a solid tool on which local decision-makers can rely to define priority protection areas.

A GIS-based DRASTIC Model and an Adjusted DRASTIC Model (DRASTICA) for Groundwater Susceptibility Assessment along the China–Pakistan Economic Corridor (CPEC) Route

Land use types and anthropogenic activities represent considerable threats to groundwater pollution. To effectively monitor the groundwater quality, it is vital to measure pollution levels before they become severe. In our research area, located in Gilgit Baltistan in northern Pakistan, groundwater supplies are diminishing due to urban sprawl. In this study, we used a GIS-based DRASTIC model (Depth to water, Recharge, Aquifer media, Soil media, Topography, Impact of the vadose zone, Hydraulic conductivity) to analyze the area’s hydrological attributes to assess the groundwater susceptibility to pollution. Considering the importance of anthropogenic activities, this research primarily utilizes an adjusted DRASTIC model called DRASTICA, which incorporates anthropogenic impact as a parameter in the model. The resulting map, which depicts vulnerability to groundwater contamination, reveals that 19% of the study area is classed as having high vulnerability, 42% has moderate vulnerability, 37% has low vulnerability, and 2% has very low vulnerability to groundwater contamination. The adopted validation process (nitrate parameter of water quality) revealed that the suggested DRASTICA model achieved better results than the established DRASTIC model in a built-up environment. We used the nitrate concentration in groundwater to verify the formulated results, and the single parameter sensitivity analysis and map removal sensitivity analysis to analyze the model sensitivity. The sensitivity analysis indicated that the groundwater vulnerability to pollution is largely influenced by anthropogenic impact and depth to the water table, thereby suggesting that anthropogenic impact must be explicitly tackled in such studies. The groundwater zones exposed to anthropogenic pollution can be better classified with the help of the proposed DRASTICA model, particularly in and around built-up environments. The responsible authorities can use this groundwater contamination data as an early warning sign, so they can take practical actions to avoid extra pressure on this vital resource.

M18. REDUCED CORTICAL CEREBRAL BLOOD FLOW IN FIRST EPISODE PSYCHOSIS PATIENTS

BackgroundAbnormal Cerebral Blood Flow (CBF) has been found in patients with chronic schizophrenia (SCZ), first-episode psychosis patients (FEP) and individuals at clinical high-risk (CHR). In particular, previous studies using Arterial Spin Labelling (ASL) found that SCZ have a global reduction of CBF in the cortex and increased CBF in the basal ganglia, the hippocampus, and the amygdala as compared with controls. To date, only one study investigated CBF using ASL in a small cohort of medicated FEP reporting increased CBF in the striatum and reduced frontal CBF as compared with controls. However, it is still not clear whether these abnormalities are related to antipsychotic treatment or rather they reflect a disease state independent from medication. Critically, clinical and pre-clinical evidence suggests that antipsychotics increase CBF, especially in the basal ganglia through dopamine D2 receptors blockade. Here, we assessed CBF differences between FEP and controls in a larger cohort of unmedicated or minimally treated patients. In addition, we tested the association between CBF abnormalities and clinical features.Methods26 FEP (13 medication-free, 9 antipsychotics naïve, 4 minimally treated) and 22 healthy controls (HC) were recruited. FEP and HC were matched for age and gender. Among FEP, 11 had a diagnosis of affective psychosis and 15 of non-affective psychosis. MRI scans were acquired using a GE MR750 3T scanner and a 12-channel head coil. ASL data was acquired using a pseudo-continuous ASL sequence (pCASL). Four control-label pairs were used. Quantification of the CBF data was made using a coil sensitivity map and a proton density image. CBF maps were normalized using non-linear registration and smoothed using a 6 mm full width at half maximum (FWHM) kernel. Grey matter CBF and regional CBF values in each Region of Interest (ROI) were extracted using individual grey matter images and WFU-Pickatlas ROIs with the MarsBar toolbox. Based on previous studies the following ROIs were selected: right and left hippocampus, right and left striatum, right and left frontal cortex. Whole-brain voxel-wise analysis was performed using a non-parametric independent t-test as implemented in FSL randomize (Threshold Free Cluster Enhancement, TFCE, and 5000 permutations). Bonferroni method was used to correct ROI analysis for multiple comparisons (alpha = 0.05, n = 7).ResultsSignificantly reduced CBF was found in FEP as compared with controls in total grey matter CBF (p = 0.004). Whole-brain voxel-wise analysis FEP-HC comparison revealed a widespread cortical reduction in large cortical clusters that encompass the occipital, parietal and frontal cortices (pFWE TFCE corrected < 0.05). ROI analysis revealed a significant reduction in CBF in the right frontal cortex (p = 0.002). A nominal significant reduction was also detected in the left frontal cortex (p = 0.014). No statistically significant differences were found in the other ROI considered (all p > 0.1). Sub-analysis after removing minimally treated patients did not change the results. Both global grey matter CBF and right frontal CBF were reduced in both affective and non-affective psychosis as compared with controls (p < 0.05), whereas no differences were detected between the two clinical groups.DiscussionOur results confirm earlier evidence on reduction of cortical perfusion in FEP. However, in contrast with previous studies, we did not detect any difference in striatal perfusion in our cohort of unmedicated/minimally treated patients. Hence, our results support the hypothesis that alteration in striatal perfusion are likely due to medication. Finally, our results suggest that CBF alterations might have a trans-diagnostic role in the pathophysiology of psychosis.

Heyelan Duyarlılık Haritalarının Üretilmesinde Analitik Hiyerarşi Yönteminin Ve Coğrafi Bilgi Sistemlerinin Kullanımı

Earthquakes and landslides are among the natural disasters that deeply affect social life. Landslide, especially occurs in mountainous areas, causes loss of life and damages every year. Studies to determine landslide sensitive areas have become quite widespread in recent years. Conducting landslide susceptibility analysis will both facilitate the planning of engineering projects and reduce the loss. In this study, landslide susceptibility maps of Saqezchi in Ardabil (Ardebil) region of Iran are created. In landslide susceptibility analysis; land use, rainfall, distance to faults, lithology, distance to river networks, elevation, slope, aspect and distance to road parameters are used. Analytical Hierarchy Process (AHP) method and Geographic Information Systems (GIS) are used while creating landslide susceptibility map. The created sensitivity maps are classified under 5 groups as “very high, high, medium, low and very low” sensitive areas.

Developing a Universal Mirror–mirror Laser Mapping System for Single Event Effect Research

Research on single event effects (SEEs) is significant to the design and manufacture of modern electronic devices. By applying two photon absorption (TPA) ultra-fast pulsed lasers, extra electron-hole pairs (EHPs) are generated in a desired location on a chip, simulating the process that could occur in the circuit by energetic particles. In this study, a SEE sensitivity mapping system is described which uses this method to generate real-time sensitivity maps for various electronic devices. The system hardware includes an attenuator to control the energy, a Pockels cell as a fast-optical switcher and a mirror–mirror module to project the laser beam into a certain location. The system software developed for this application controls the laser system, automatically generates sensitivity maps, communicates with the testing devices and logs the SEE results. The two main features of this laser mapping system are: high scanning velocity for large area scanning (about 1 × 1 mm) and high spatial resolution for small area scanning (about 1 × 1 μm). To verify this mapping system, sensitivity maps were generated for static random access memory (SRAM) built with 65 nm technology and for commercial operational amplifiers (op-amps). The achieved sensitivity maps were compared with circuitry analysis and laser testing results, confirming this mapping system to be effective.

WITHDRAWN: Environmental dimensionality determines species coexistence

According to the competitive-exclusion principle, the number n of regulating variables describing a given community dynamics is an upper bound on the number of species (or types or morphs) that can coexist at equilibrium. On occasion, it is possible to reformulate a model with a lower number of regulating variables than appeared in the initial specification. We call the smallest number of such variables the dimension of the environmental feedback, or environmental dimension for short. For studying which species can invade a community, it is enough to know the sign of each species' long-term growth rate, i.e., invasion fitness. Therefore, different indicators of population growth - so-called fitness proxies, such as the basic reproduction number-are sometimes preferred. However, as we show, different fitness proxies may have different dimensions. Fundamental characteristics such as the environmental dimension should not depend on such arbitrary choices. Here, we resolve this difficulty by introducing a refined definition of environmental dimension that focuses on neutral fitness contours. On this basis, we show that this definition of environmental dimension is not only unambiguous, i.e., independent of the choice of fitness proxy, but also constructive, i.e., applicable without needing to assess an infinite number of possible fitness proxies. We then investigate how to determine environmental dimensions by analysing the two components of the environmental feedback: the impact map describing how a community's resident species affect the regulating variables and the sensitivity map describing how population growth depends on the regulating variables. The dimension of the impact map is lower than n when the set of feasible environments is of lower dimension than n, and the dimension of the sensitivity map is lower than n when not all n regulating variables affect the sign of population growth independently. While the minimum of the dimensions of the impact and sensitivity maps provides an upper bound on the environmental dimension, the combined effect of the two maps can result in an even lower environmental dimension, which happens when the sensitivity map is insensitive to some aspects of the impact map's image. To facilitate the applications of the framework introduced here, we illustrate all key concepts with detailed worked examples. In view of these results, we claim that the environmental dimension is the ultimate generalization of the traditional and widely used notions of the "number of regulating variables" or the "number of limiting factors", and is thus the sharpest generally applicable upper bound on the number of species that can robustly coexist in a community.

Artificial intelligence for early prediction of pulmonary hypertension using electrocardiography

Screening and early diagnosis of pulmonary hypertension (PH) are critical for managing progression and preventing associated mortality; however, there are no tools for this purpose. We developed and validated an artificial intelligence (AI) algorithm for predicting PH using electrocardiography (ECG). This historical cohort study included data from consecutive patients from 2 hospitals. The patients in one hospital were divided into derivation (56,670 ECGs from 24,202 patients) and internal validation (3,174 ECGs from 3,174 patients) datasets, whereas the patients in the other hospital were included in only an external validation (10,865 ECGs from 10,865 patients) dataset. An AI algorithm based on an ensemble neural network was developed using 12-lead ECG signal and demographic information from the derivation dataset. The end-point was the diagnosis of PH. In addition, the interpretable AI algorithm identified which region had the most significant effect on decision making using a sensitivity map. During the internal and external validation, the area under the receiver operating characteristic curve of the AI algorithm for detecting PH was 0.859 and 0.902, respectively. In the 2,939 individuals without PH at initial echocardiography, those patients that the AI defined as having a higher risk had a significantly higher chance of developing PH than those in the low-risk group (31.5% vs 5.9%, p < 0.001) during the follow-up period. The sensitivity map showed that the AI algorithm focused on the S-wave, P-wave, and T-wave for each patient by QRS complex characteristics. The AI algorithm demonstrated high accuracy for PH prediction using 12-lead and single-lead ECGs.

Sensitivity maps for time-reverse imaging: an accuracy study for the Los Humeros Geothermal Field (Mexico)

SUMMARY The estimation of the source–location accuracy of microseismic events in reservoirs is of significant importance. Time-reverse imaging (TRI) provides a highly accurate localization scheme to locate events by time-reversing the recorded full wavefield and back propagating it through a velocity model. So far, the influence of the station geometry and the velocity model on the source–location accuracy is not well known. Therefore, sensitivity maps are developed using the geothermal site of Los Humeros in Mexico to evaluate the spatial variability of the source–location accuracy. Sensitivity maps are created with an assumed gradient velocity model with a constant vp–vs ratio and with a realistic velocity model for the region of Los Humeros. The positions of 27 stations deployed in Los Humeros from September 2017 to September 2018 are used as surface receivers. An automatic localization scheme is proposed that does not rely on any a priori information about the sources and thus negates any user bias in the source locations. The sensitivity maps are created by simulating numerous uniformly distributed sources simultaneously and locating these sources using TRI. The found source locations are compared to the initial source locations to estimate the achieved accuracy. The resulting sensitivity maps show that the station geometry introduces complex patterns in the spatial variation of accuracy. Furthermore, the influence of the station geometry on the source–location accuracy is larger than the influence of the velocity model. Finally, a microearthquake recorded at the geothermal site of Los Humeros is located to demonstrate the usability of the derived sensitivity maps. This study stresses the importance of optimizing station networks to enhance the accuracy when locating seismic events using TRI.

Steady-state response of annular wedge-shaped aquifers to arbitrarily-located multiwells with regional flow

Wedge-shaped aquifers are sometimes cut by crossing streams so that the areal geometry can be delineated by an annular wedge of infinite radial extent. Despite their worldwide spread, the problem of groundwater flow in such geometrical entities has not received much attention in analytical studies. This paper presents a closed-form analytical solution for the aquifer head distribution in response to a system of multiwells superimposed on a background regional flow. The aquifer is isotropic, homogeneous and in prefect hydraulic connection with the surrounding streams. A steady-state Green’s function is constructed for this laterally bounded aquifer system. The formulation is general in the sense that the aquifer vertex angle as well as number of wells and their positions can be arbitrary chosen. Closed-form expression for the stream function is obtained by integration of the Cauchy-Riemann equations. Three types of hydrogeological boundary conditions are considered, combining streams of constant and linearly varying head and no-flow barrier. The results are applicable to both confined and unconfined flows. Extremely simple expressions are derived to quantify the amount of water exchange across stream-aquifer interfaces. Flow nets are given for a number of hypothetical test cases, demonstrating different aspects of the generated flow field. Sensitivity maps are prepared for the stream depletion rates due to a single pumping well and the role of aquifer vertex angle on the model response is investigated. Multiple stagnation points in the flow field are identified on a semi-analytical basis. This helps to assess how the position of a single stagnation point is affected by changes in the gradient of regional flow. For relatively high gradients, it is observed that the radial position of the stagnation point coincides with that of the well itself. Finally, the formulation is extended to account for rainfall recharge over an unconfined aquifer having the same areal extent. The aquifer response is found to agree well with other analytical and numerical results.

The potential sensitivity to climate change of selected endangered and important Natura 2000 Habitats and plants from Bucegi Natural Park, Romania

This study was carried out in the Bucegi Natural Park, a protected area of the Romanian Carpathians. It aims at documenting the potential sensitivity of six widespread Natura 2000 habitat types and of all plants with conservative value (200 taxa) in the mountain area, to the changes in temperature and humidity, predicted for this century. Regional expert knowledge and environmental indicator values were considered in assessing the potential habitat’s sensitivity. The results support the evidence that sensitivity to temperature may be potentially higher for habitats at alpine and subalpine levels (bushes and grasslands) and medium for forest habitats. Sensitivity to moisture was detected as potentially high for forest habitats and as medium for bushes and grasslands at high mountain elevation. Microthermophilic plants have shown a greater share (76-79%) in alpine and subalpine communities, and the hydrophilic plants (86-96%) in forest communities. About 80% of plants of conservation value (microthermophilic or hydrophilic plants) may be potentially sensitive to predicted warming and drought and 44% of them (microthermophilic and hydrophilic plants) to the changes of both parameters. Climate scenarios (2011-2100) and sensitivity maps (Sat – image interpretation with GIS for the whole mountain area) are included.

Seed testing of foliar-fertilised red clover crops after various periods of storage

This study was carried out in the Bucegi Natural Park, a protected area of the Romanian Carpathians. It aims at documenting the potential sensitivity of six widespread Natura 2000 habitat types and of all plants with conservative value (200 taxa) in the mountain area, to the changes in temperature and humidity, predicted for this century. Regional expert knowledge and environmental indicator values were considered in assessing the potential habitat’s sensitivity. The results support the evidence that sensitivity to temperature may be potentially higher for habitats at alpine and subalpine levels (bushes and grasslands) and medium for forest habitats. Sensitivity to moisture was detected as potentially high for forest habitats and as medium for bushes and grasslands at high mountain elevation. Microthermophilic plants have shown a greater share (76-79%) in alpine and subalpine communities, and the hydrophilic plants (86-96%) in forest communities. About 80% of plants of conservation value (microthermophilic or hydrophilic plants) may be potentially sensitive to predicted warming and drought and 44% of them (microthermophilic and hydrophilic plants) to the changes of both parameters. Climate scenarios (2011-2100) and sensitivity maps (Sat – image interpretation with GIS for the whole mountain area) are included.

Artificial Intelligence for Early Prediction of Pulmonary Hypertension Using Electrocardiography

Although screening and early diagnosis is critical for preventing irreversible progression of pulmonary hypertension (PH) and preventable mortality, there are no suitable tools for this purpose. We developed and validated an artificial intelligence (AI) algorithm for predicting PH using electrocardiography (ECG). This retrospective cohort study included data from two hospitals. An AI algorithm was developed using 56,670 ECGs from 24,202 patients. Internal validation of the algorithm was performed with 3174 patients from one hospital, while external validation was performed with 10,865 patients from another hospital. The endpoint was the diagnosis of PH, confirmed by echocardiography. We used demographic information, features, and 500-Hz ECG raw data as predictive variables. Additionally, we identified which region had the most significant effect on the decision-making of the AI algorithm using a sensitivity map. During the internal and external validation, the area under the receiver operating characteristic curve of the AI algorithm using a 12-lead ECG for detecting PH was 0.859 and 0.902, respectively, while that using a single-lead ECG was 0.824 and 0.832, respectively. In the 2939 non-PH individuals at initial echocardiography, those patients that the AI defined as high risk had a significantly higher chance of development of PH than the low risk group (31.5% vs. 5.9%, p <0.001) during the follow-up period. The sensitivity map showed the AI algorithm focused on the S-wave, P-wave, and T-wave for each patient by QRS complex characteristics. The AI algorithm demonstrated high accuracy for PH prediction using 12-lead and single-lead ECGs.

Deep Learning-Based Algorithm for Detecting Aortic Stenosis Using Electrocardiography.

BackgroundSevere, symptomatic aortic stenosis (AS) is associated with poor prognoses. However, early detection of AS is difficult because of the long asymptomatic period experienced by many patients, during which screening tools are ineffective. The aim of this study was to develop and validate a deep learning–based algorithm, combining a multilayer perceptron and convolutional neural network, for detecting significant AS using ECGs.Methods and ResultsThis retrospective cohort study included adult patients who had undergone both ECG and echocardiography. A deep learning–based algorithm was developed using 39 371 ECGs. Internal validation of the algorithm was performed with 6453 ECGs from one hospital, and external validation was performed with 10 865 ECGs from another hospital. The end point was significant AS (beyond moderate). We used demographic information, features, and 500‐Hz, 12‐lead ECG raw data as predictive variables. In addition, we identified which region had the most significant effect on the decision‐making of the algorithm using a sensitivity map. During internal and external validation, the areas under the receiver operating characteristic curve of the deep learning–based algorithm using 12‐lead ECG for detecting significant AS were 0.884 (95% CI, 0.880–0.887) and 0.861 (95% CI, 0.858–0.863), respectively; those using a single‐lead ECG signal were 0.845 (95% CI, 0.841–0.848) and 0.821 (95% CI, 0.816–0.825), respectively. The sensitivity map showed the algorithm focused on the T wave of the precordial lead to determine the presence of significant AS.ConclusionsThe deep learning–based algorithm demonstrated high accuracy for significant AS detection using both 12‐lead and single‐lead ECGs.

Crossing Domains for AU Coding: Perspectives, Approaches, and Measures.

Facial action unit (AU) detectors have performed well when trained and tested within the same domain. How well do AU detectors transfer to domains in which they have not been trained? We review literature on cross-domain transfer and conduct experiments to address limitations of prior research. We evaluate generalizability in four publicly available databases. EB+ (an expanded version of BP4D+), Sayette GFT, DISFA and UNBC Shoulder Pain (SP). The databases differ in observational scenarios, context, participant diversity, range of head pose, video resolution, and AU base rates. In most cases performance decreased with change in domain, often to below the threshold needed for behavioral research. However, exceptions were noted. Deep and shallow approaches generally performed similarly and average results were slightly better for deep model compared to shallow one. Occlusion sensitivity maps revealed that local specificity was greater for AU detection within than cross domains. The findings suggest that more varied domains and deep learning approaches may be better suited for generalizability and suggest the need for more attention to characteristics that vary between domains. Until further improvement is realized, caution is warranted when applying AU classifiers from one domain to another.

Detectability of cerebellar activity with magnetoencephalography and electroencephalography.

Electrophysiological signals from the cerebellum have traditionally been viewed as inaccessible to magnetoencephalography (MEG) and electroencephalography (EEG). Here, we challenge this position by investigating the ability of MEG and EEG to detect cerebellar activity using a model that employs a high‐resolution tessellation of the cerebellar cortex. The tessellation was constructed from repetitive high‐field (9.4T) structural magnetic resonance imaging (MRI) of an ex vivo human cerebellum. A boundary‐element forward model was then used to simulate the M/EEG signals resulting from neural activity in the cerebellar cortex. Despite significant signal cancelation due to the highly convoluted cerebellar cortex, we found that the cerebellar signal was on average only 30–60% weaker than the cortical signal. We also made detailed M/EEG sensitivity maps and found that MEG and EEG have highly complementary sensitivity distributions over the cerebellar cortex. Based on previous fMRI studies combined with our M/EEG sensitivity maps, we discuss experimental paradigms that are likely to offer high M/EEG sensitivity to cerebellar activity. Taken together, these results show that cerebellar activity should be clearly detectable by current M/EEG systems with an appropriate experimental setup.

Study on hot deformation behaviour and microstructure of Zr-4 alloy

The isothermal compression experiment of Zr-4 alloy was accomplished by using Gleeble-3500 thermomechanical simulator. The hot deformation behaviour at the strain rate of 0.001~10s-1 and the deformation temperature of 750~1000°C was researched. The characteristics of flow stress were studied. The result is that the flow stress has a great influence on the deformation temperature and the strain rate. Based on the experimental data, the strain rate sensitivity map (m) was established. Through microstructure observation, the zones at lower m value are mainly flow localization (750 °C/10s-1) and kink (950 °C/10s-1). The zones at higher m value are mainly spheroidization. The result is that at strain rate of 0.01~0.1s-1/deformation temperature of 900~1000 °C is the suitable deformation parameter of Zr-4 alloy.