Search

BackgroundCOVID-19 has infected millions of people worldwide and is responsible for several hundred thousand fatalities. The COVID-19 pandemic has necessitated thoughtful resource allocation and early identification of high-risk patients. However, effective methods to meet these needs are lacking. ObjectiveThe aims of this study were to analyze the electronic health records (EHRs) of patients who tested positive for COVID-19 and were admitted to hospitals in the Mount Sinai Health System in New York City; to develop machine learning models for making predictions about the hospital course of the patients over clinically meaningful time horizons based on patient characteristics at admission; and to assess the performance of these models at multiple hospitals and time points. MethodsWe used Extreme Gradient Boosting (XGBoost) and baseline comparator models to predict in-hospital mortality and critical events at time windows of 3, 5, 7, and 10 days from admission. Our study population included harmonized EHR data from five hospitals in New York City for 4098 COVID-19–positive patients admitted from March 15 to May 22, 2020. The models were first trained on patients from a single hospital (n=1514) before or on May 1, externally validated on patients from four other hospitals (n=2201) before or on May 1, and prospectively validated on all patients after May 1 (n=383). Finally, we established model interpretability to identify and rank variables that drive model predictions. ResultsUpon cross-validation, the XGBoost classifier outperformed baseline models, with an area under the receiver operating characteristic curve (AUC-ROC) for mortality of 0.89 at 3 days, 0.85 at 5 and 7 days, and 0.84 at 10 days. XGBoost also performed well for critical event prediction, with an AUC-ROC of 0.80 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. In external validation, XGBoost achieved an AUC-ROC of 0.88 at 3 days, 0.86 at 5 days, 0.86 at 7 days, and 0.84 at 10 days for mortality prediction. Similarly, the unimputed XGBoost model achieved an AUC-ROC of 0.78 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. Trends in performance on prospective validation sets were similar. At 7 days, acute kidney injury on admission, elevated LDH, tachypnea, and hyperglycemia were the strongest drivers of critical event prediction, while higher age, anion gap, and C-reactive protein were the strongest drivers of mortality prediction. ConclusionsWe externally and prospectively trained and validated machine learning models for mortality and critical events for patients with COVID-19 at different time horizons. These models identified at-risk patients and uncovered underlying relationships that predicted outcomes.

Key Clinical Message: Consideration of spontaneous urinary bladder rupture in the differential diagnosis of acute abdominal pain for alcohol‐abusing patients is crucial for ensuring timely surgical intervention and preventing life‐threatening complications due to its high associated morbidity and mortality. Spontaneous rupture of the urinary bladder (SRUB) is a rare but critical urological emergency, typically associated with malignancy, neurogenic dysfunction, or previous radiation therapy. Here, we present a unique case of SRUB in a 65‐year‐old chronic alcoholic male who presented with acute lower abdominal pain following heavy alcohol consumption. Initial evaluations revealed leukocytosis, elevated serum creatinine levels, and ultrasound findings suggestive of bladder rupture. Computed tomography confirmed the diagnosis, indicating an intraperitoneal rupture with associated hematoma. Immediate surgical repair was performed, leading to a successful outcome. This case underscores the importance of considering SRUB in patients with acute abdominal pain, especially in the context of alcohol intoxication, and highlights the diagnostic and therapeutic challenges associated with this condition. Early recognition and intervention are crucial to prevent life‐threatening complications associated with urinary bladder rupture. [ABSTRACT FROM AUTHOR]

Groundwater contamination due to the leaching of harmful pollutants such as heavy metals, xenobiotic compounds, and other inorganic compounds from solid waste dumping sites has become a major health concern in recent times. Therefore, to assess the effects of the Bandhwari landfill site, groundwater samples from the surrounding region of the dumping site were collected and analyzed for heavy metals and physicochemical properties. The total dissolved solids (TDS) of 67% of samples exceeded the Bureau of Indian Standards (BIS) permissible limits which makes it unfit for drinking purposes. The groundwater samples were also analyzed for iron (Fe), lead (Pb), zinc (Zn), nickel (Ni), copper (Cu), cadmium (Cd), and chromium (Cr) concentrations and results of heavy metal concentration in the groundwater around the Bandhwari landfill follow the concentration trend of Pb.Cd.Ni.Cu.Zn.Fe.Cr. Risk assessment of consumers' health was done using target hazard quotient calculations which were less than unity (threshold value of ,1), indicating that heavy metal concentrations do not pose any serious health effect according to total hazard quotient values. The results of the study made it evident that groundwater is not suitable for drinking purposes due to excess values of water quality parameters but poses no risk due to studied metal concentrations. [ABSTRACT FROM AUTHOR]

In this paper, we are interested in soft tissue differentiation by multiband images obtained from the High-Resolution Ultrasonic Transmission Tomography (HUTT) system using a spectral target detection method based on constrained energy minimization (CEM). We have developed a new tissue differentiation method (called 'CEM filter bank') consisting of multiple CEM filters specially designed for detecting multiple types of tissues. Statistical inference on the output of the CEM filter bank is used to make a decision based on the maximum statistical significance rather than the magnitude of each CEM filter output. We test and validate this method through three-dimensional interphantom/intraphantom soft tissue classification where target profiles obtained from an arbitrary single slice are used for differentiation over multiple other tomographic slices. The performance of the proposed classifier is assessed using receiver operating characteristic analysis. We also apply our method to classify tiny structures inside a bovine kidney and sheep kidneys. Using the proposed method we can detect physical objects and biological tissues such as styrofoam balls, chicken tissue, calyces, and vessel-duct successfully. Index Terms--Constrained energy minimization (CEM), filter bank, high-resolution ultrasound transmission tomography (HUTT), multiband signatures, soft tissue differentiation.

Determination of fiber orientation from diffusion tensor is problematic when the tensor is not linear anisotropic. Particularly planar anisotropy is often an indicator of multiple fibers in a voxel. A novel method has been developed to identify the orientations of multiple fiber tracts in a voxel of diffusion tensor magnetic resonance imaging (MRI) using independent component analysis (ICA). A computationally efficient algorithm to estimate the independent sources has been derived by introducing a new adaptive nonlinear function to model the cumulative distribution of the sources. Monte Carlo simulation was used to evaluate the method. Simulations suggest that the orientations of two tensors in a voxel can be estimated with mean error of less than 10[degrees] for most interfiber angles when the signal-to-noise ratio is higher than 30. A processing and source selection strategy has been proposed and successfully tested with simulated tensor fields incorporating fiber crossing and with human data. Qualitative assessment of the result from human data analysis demonstrated that the ICA method reasonably estimated multiple fiber orientations corresponding to anatomically known white matter tracts. Index Terms--Diffusion-tensor magnetic resonance imaging (MRI), independent component analysis, Monte Carlo simulation, multiple fiber orientations.

The influence of conductivity tensor on the forward solution of electroencephalography was assessed in 2-D head models of a human subject. The conductivity tensors of different regions of the head were estimated from magnetic resonance-diffusion tensor images by linearly mapping the mean trace values to the published conductivity values. The anisotropic conductivity model was compared with the isotropic conductivity model in terms of the difference between the scalp potentials. The differences were measured by the cross correlation (CC) and the relative error (RE) between two models. We have also proposed a new measure, scaling-removed RE (SRRE) as a more effective indicator of the difference. The results with 354 individual dipole sources show that there are remarkable differences between the anisotropic conductivity tensor and the isotropic model (CC = 0.96, RE = 30.73% and SRRE = 19.34%). Although the CC is high, the large RE and SRRE indicate that this difference may also affect the accuracy of inverse solutions in localizing the current dipole sources. Index Terms--Conductivity tensor, diffusion tensor-magnetic resonance images (DT-MRI), electroencephalography (EEG), finite-element method (FEM).

The temporal resolution of functional MRI (fMRI) is limited by the shape of the hemodynamic response function (hrf) and the vascular architecture underlying the activated regions. Typically, the temporal resolution of fMRI is on the order of 1 s. We have developed a new data processing approach to extract temporal information on a pixel-by-pixel basis at the level of 100 ms from fMRI data. Instead of correlating or fitting the time-course of each pixel to a single reference function, which is the common practice in fMRI, we correlate each pixel's time-course to a series of reference functions that are shifted with respect to each other by 100 ms. The reference function yielding the highest correlation coefficient for a pixel is then used as a time marker for that pixel. A Monte Carlo simulation and experimental study of this approach were performed to estimate the temporal resolution as a function of signal-to-noise ratio (SNR) in the time-course of a pixel. Assuming a known and stationary hrf, the simulation and experimental studies suggest a lower limit in the temporal resolution of approximately 100 ms at an SNR of 3. The multireference function approach was also applied to extract timing information from an event-related motor movement study where the subjects flexed a finger on cue. The event was repeated 19 times with the event's presentation staggered to yield an approximately 100-ms temporal sampling of the hemodynamic response over the entire presentation cycle. The timing differences among different regions of the brain activated by the motor task were clearly visualized and quantified by this method. The results suggest that it is possible to achieve a temporal resolution of ~200 ms in practice with this approach. Index Terms--Functional MRI, motor activity, temporal resolution.

We have investigated electroencephalography (EEG) distributed source imaging with a realistic finite-element (FE) head model. The performance of different FE imaging methods was evaluated and compared in two- (2-D) and three-dimensional (3-D) simulation studies. The results demonstrate the feasibility of EEG distributed source imaging with FE head models using FE inverse methods. We also show that incorporating prior knowledge of sources significantly improves the inverse solutions. As an application of the EEG imaging methods, the sources of voluntary movement-related evoked potentials of the human brain were imaged. The reconstructed EEG sources show good spatial correlation with the functional magnetic imaging resonance (fMRI)-mapped motor activity in the brain, thus validating the FE inverse methods and adding temporal information to the fMRI-mapped regions of the brain. Index Terms--Electroencephalography (EEG) source imaging, finite-element (FE) inverse methods, finite-element realistic head modeling, functional magnetic resonance imaging (fMRI).

A high-resolution SPECT instrument dedicated to breast imaging has been designed incorporating arrays of collimated Cadmium-Zinc-Telluride (CZT) detectors tiled on either a cylindrical surface or a hemispherical surface surrounding the breast. The performance characteristics of a three-segment rotating parallel-hole collimator as well as a stationary multiple pin-hole collimator were considered for the cylindrical system. A stationary multiple pin-hole collimated system was also considered for the hemispherical design. Monte Carlo studies suggest that at almost equal spatial resolution of 0.5cm, the cylindrical design with parallel-hole collimator would have an approximately a factor of two higher geometrical efficiency than the hemispherical pinhole collimated system including effects of attenuation in the breast. However, whereas the parallel-hole collimator must be rotated to acquire data from multiple angles of view, the pin-hole version has the advantage of recording data from 112 views in a stationary mode. Monte Carlo studies of filtered backprojection as well as a Bayesian reconstruction approach including attenuation and scatter within the breast, where the breast was modeled as a 15cm hemisphere of uniform activity distribution containing three spherical lesions of diameters 1.0cm, 0.7cm and 0.5cm respectively suggest that the 0.5cm could be detected with either design in a one-hour SPECT study assuming a 10:1 tumor to background ratio. We conclude that a high resolution Breast SPECT instrument where the resolution is limited to about 0.5cm is viable with CZT detectors.

The estimation of multiple dipole parameters in spatio-temporal source modeling (STSM) of electroencephalographic (EEG) data is a difficult nonlinear optimization problem due to multiple local minima in the cost function. A straightforward iterative optimization approach to such a problem is very susceptible to being trapped in a local minimum, thereby resulting in incorrect estimates of the dipole parameters. In this paper, we present and evaluate a more robust optimization approach based on the simulated annealing algorithm. The complexity of this approach for the STSM problem was reduced by separating the dipole parameters into linear (moment) and nonlinear (location) components. The effectiveness of the proposed method and its superiority over the traditional nonlinear simplex technique in escaping local minima were tested and demonstrated through computer simulations. The annealing algorithm and its implementation for multidipole estimation are also discussed. We found the simulated annealing approach to be 7-31% more effective than the simplex method at converging to the true global minimum for a number of different kinds of three-dipole problems simulated in this work. In addition, the computational cost of the proposed approach was only marginally higher than its simplex counterpart. The annealing method also yielded similar solutions irrespective of the initial guesses used. The proposed simulated annealing method is an attractive alternative to the simplex method that is currently more common in dipole estimation applications. Index Terms - Brain mapping, dipoles, electroencephalography, estimation, functional imaging, human, inverse problem, nonlinear optimization.

An accurate technique has been developed to correct the intensity variation in the MR images acquired with an endorectal surface coil for prostate imaging. The technique relies on reconstructing the in-vivo size, shape and location of the coil during imaging, wherefrom a pixel-by-pixel correction is applied to the coil's sensitivity profile in accordance with the Biot-Savart law. The method has been evaluated by a test-object containing multiple layers of beads embedded in gelatine. Qualitatively, the improvement in visualizing regions of interest after corrections is obvious. Quantitatively, the local snr shows an improvement by a factor of 1.5-2. Several human studies have also been performed, and the technique shows remarkable improvement in visualizing the prostate region.

Studies were conducted to detect and demonstrate the increase in lactate in the human auditory cortex during auditory stimulation. The studies were hoped to facilitate the use of lactate activity in spectroscopic imaging of stimulated brain function. However, there is inter-subject variability in the time activity and increase in lactate during stimulation which entails further research.