Development of a machine learning-based clinical decision support system to predict clinical deterioration in patients visiting the emergency department.

This study aimed to develop a machine learning-based clinical decision support system for emergency departments based on the decision-making framework of physicians. We extracted 27 fixed and 93 observation features using data on vital signs, mental status, laboratory results, and electrocardiograms during emergency department stay. Outcomes included intubation, admission to the intensive care unit, inotrope or vasopressor administration, and in-hospital cardiac arrest. eXtreme gradient boosting algorithm was used to learn and predict each outcome. Specificity, sensitivity, precision, F1 score, area under the receiver operating characteristic curve (AUROC), and area under the precision-recall curve were assessed. We analyzed 303,345 patients with 4,787,121 input data, resampled into 24,148,958 1 h-units. The models displayed a discriminative ability to predict outcomes (AUROC > 0.9), and the model with lagging 6 and leading 0 displayed the highest value. The AUROC curve of in-hospital cardiac arrest had the smallest change, with increased lagging for all outcomes. With inotropic use, intubation, and intensive care unit admission, the range of AUROC curve change with the leading 6 was the highest according to different amounts of previous information (lagging). In this study, a human-centered approach to emulate the clinical decision-making process of emergency physicians has been adopted to enhance the use of the system. Machine learning-based clinical decision support systems customized according to clinical situations can help improve the quality of care.

Dyeing of Polyester with 4-Fluorosulfonylphenylazo-5-pyrazolone Disperse Dyes and Application of Environment-Friendly Aftertreatment for Their High Color Fastness.

Dyeing and fastness properties of a series of 4-fluorosulfonylphenylazo-5-pyrazolone dyes on polyester were investigated in this study. The 4-nitrophenylazo-5-pyrazolone dyes were also synthesized to compare their dyeing and fastness properties on polyester with those of fluorosulfonyl-substituted analogues. The substantivity of 4-arylazo-5-pyrazolone derivatives containing a p-fluorosulfonyl group in the diazo component was lower than that of their nitro analogues which have a higher extinction coefficient and higher affinity because of the polar nitro group. They showed relatively hypsochromic color and lower chroma on polyester compared with their nitro analogues because of the relatively weaker electron-accepting power of the fluorosulfonyl group compared to the nitro group. Disperse dyeing of polyester with 4-fluorosulfonylphenylazo-5-pyrazolone disperse dyes achieved high color fastness and reduces the adverse environmental impact of the dyeing process by providing the option of performing alkali clearing instead of reductive clearing, which has high biological oxygen demand when discharged into the dyeing effluent and generates carcinogenic aromatic amines.

Robust sleep quality quantification method for a personal handheld device.

OBJECTIVE: The purpose of this study was to develop and validate a novel method for sleep quality quantification using personal handheld devices. MATERIALS AND METHODS: The proposed method used 3- or 6-axes signals, including acceleration and angular velocity, obtained from built-in sensors in a smartphone and applied a real-time wavelet denoising technique to minimize the nonstationary noise. Sleep or wake status was decided on each axis, and the totals were finally summed to calculate sleep efficiency (SE), regarded as sleep quality in general. The sleep experiment was carried out for performance evaluation of the proposed method, and 14 subjects participated. An experimental protocol was designed for comparative analysis. The activity during sleep was recorded not only by the proposed method but also by well-known commercial applications simultaneously; moreover, activity was recorded on different mattresses and locations to verify the reliability in practical use. Every calculated SE was compared with the SE of a clinically certified medical device, the Philips (Amsterdam, The Netherlands) Actiwatch. RESULTS: In these experiments, the proposed method proved its reliability in quantifying sleep quality. Compared with the Actiwatch, accuracy and average bias error of SE calculated by the proposed method were 96.50% and -1.91%, respectively. CONCLUSIONS: The proposed method was vastly superior to other comparative applications with at least 11.41% in average accuracy and at least 6.10% in average bias; average accuracy and average absolute bias error of comparative applications were 76.33% and 17.52%, respectively.

Characteristic time scales of electroencephalograms of narcoleptic patients and healthy controls.

Sleep electroencephalograms (EEGs) typically showed correlated fluctuations that became random-like oscillations beyond a characteristic time scale. To investigate this behavior quantitatively, the detrended fluctuation analysis (DFA) was applied to EEGs of 10 narcoleptic patients (22.0 ± 4.0 yrs; 6 males) and 8 healthy controls (24.0 ± 2.0 yrs; 5 males). The characteristic time scales of the narcoleptics and controls were estimated as 1.8 ± 0.7 and 4.4 ± 1.2s, respectively (significance level, p<0.01). We further performed DFA of the EEGs segmented into 30s epochs and found that the DFA scaling exponents increased in deep sleep stages. These results were verified with power spectrum and auto-correlation analysis, and reproduced by a mathematical model. We thus concluded that characteristics of EEGs of narcoleptic patients could be differentiated from those of healthy subjects, suggesting a potential application of DFA in diagnosing narcolepsy.

Robust measurement of ocular torsion using iterative Lucas-Kanade.

We present a new method to measure ocular torsion using Lucas-Kanade method. After pixels of iris annulus around a pupil have been converted into Cartesian coordinates, 30 features on the iris was selected then the features were tracked using the iterative Lucas-Kanade algorithm to calculate torsional shift. The results show that a precision of the method is higher than those measured by a conventional cross-correlation and by a template matching method. The suggested method showed 0.03 degrees mean error with 0.15 degrees maximum error. Particularly, the method was robust to change of pupil size and misalignment of pupil location. Processing time was also fast enough to be implemented in a real-time system.

The development of a nonintrusive home-based physiologic signal measurement system.

This paper describes the development of a nonintrusive health-monitoring house system that has been developed for monitoring patients' health status. The system functions unobtrusively and leaves patients free to conduct daily activities. Installed sensors and devices in the house monitor biosignals such as bedside electrocardiogram (ECG), snoring, weight, and movement pattern. The network in the house was constructed using asymmetric digital subscriber line (ADSL) and Bluetooth technologies and it is capable of transmitting monitored signals to the remote hospital server through the home server. The system is found to be effective in monitoring patients' daily activity and health status in a nonconstraining manner.

Development of pupillography using image processing.

PURPOSE: Pupillary examination is an important objective method to diagnose lesions of the anterior visual pathways. However, errors and faults may easily alter the interpretation and value of the test as it is highly dependent on the examiner's skills. Therefore, we tried to develop a pupillography which is independent of the examiner. METHODS: Hardware composed of a binocularly measuring instrument adapted for an infrared charge coupled device (CCD) was developed. Two arrays of infrared light emitting diodes (LED) were supplied in front of each of the subject's eyes. A microcontroller to modulate these LED was developed, as was software to save and analyze the pupil images. The hardware was able to deliver a light to either eye or to both eyes, and to change the time, frequency, and intensity of the stimulus. The software automatically analyzed the pupil size and location by image processing. Pupil size was calculated continuously. After artifact elimination, the response amplitudes of the pupils were determined for the right and left pupils. RESULTS: Pupillary images of size 320 x 240, at 30 frames/second, were saved and processed to evaluate the change of the actual pupil size and the velocity of pupillary response. CONCLUSIONS: A pupillography to measure, save and analyze the pupillary response using image processing was developed. Further detailed clinical studies with a large number of patients will be required to validate this new method.