Data pipeline for real-time energy consumption data management and prediction.

With the increasing utilization of data in various industries and applications, constructing an efficient data pipeline has become crucial. In this study, we propose a machine learning operations-centric data pipeline specifically designed for an energy consumption management system. This pipeline seamlessly integrates the machine learning model with real-time data management and prediction capabilities. The overall architecture of our proposed pipeline comprises several key components, including Kafka, InfluxDB, Telegraf, Zookeeper, and Grafana. To enable accurate energy consumption predictions, we adopt two time-series prediction models, long short-term memory (LSTM), and seasonal autoregressive integrated moving average (SARIMA). Our analysis reveals a clear trade-off between speed and accuracy, where SARIMA exhibits faster model learning time while LSTM outperforms SARIMA in prediction accuracy. To validate the effectiveness of our pipeline, we measure the overall processing time by optimizing the configuration of Telegraf, which directly impacts the load in the pipeline. The results are promising, as our pipeline achieves an average end-to-end processing time of only 0.39 s for handling 10,000 data records and an impressive 1.26 s when scaling up to 100,000 records. This indicates 30.69-90.88 times faster processing compared to the existing Python-based approach. Additionally, when the number of records increases by ten times, the increased overhead is reduced by 3.07 times. This verifies that the proposed pipeline exhibits an efficient and scalable structure suitable for real-time environments.

Construction Site Safety Management: A Computer Vision and Deep Learning Approach.

In this study, we used image recognition technology to explore different ways to improve the safety of construction workers. Three object recognition scenarios were designed for safety at a construction site, and a corresponding object recognition model was developed for each scenario. The first object recognition model checks whether there are construction workers at the site. The second object recognition model assesses the risk of falling (falling off a structure or falling down) when working at an elevated position. The third object recognition model determines whether the workers are appropriately wearing safety helmets and vests. These three models were newly created using the image data collected from the construction sites and synthetic image data collected from the virtual environment based on transfer learning. In particular, we verified an artificial intelligence model based on a virtual environment in this study. Thus, simulating and performing tests on worker falls and fall injuries, which are difficult to re-enact by humans, are efficient algorithm verification methods. The verification and synthesis data acquisition method based on a virtual environment is one of the main contributions of this study. This paper describes the overall application development approach, including the structure and method used to collect the construction site image data, structure of the training image dataset, image dataset augmentation method, and the artificial intelligence backbone model applied for transfer learning.

Applicability of a Combined DAF-MF Process to Respond to Changes in Reservoir Water Quality through a Two-Year Pilot Plant Operation.

The optimal operating conditions of a combined dissolved air flotation (DAF)-microfiltration (MF) process to respond to changes in raw water quality were investigated by operating a pilot plant for two years. Without DAF pre-treatment (i.e., MF alone), MF operated stably with a transmembrane pressure (TMP) increase of 0.24 kPa/d when the turbidity of raw water was low and stable (max. 13.4 NTU). However, as the raw water quality deteriorated (max. 76.9 NTU), the rate of TMP increase reached 43.5 kPa/d. When DAF pre-treatment was applied (i.e., the combined DAF-MF process), the MF process operated somewhat stably; however, the rate of TMP increase was relatively high (i.e., 0.64 kPa/d). Residual coagulants and small flocs were not efficiently separated by the DAF process, exacerbating membrane fouling. Based on the particle count analysis of the DAF effluent, the DAF process was optimised based on the coagulant dose and hydraulic loading rate. After optimisation, the rate of TMP increase for the MF process stabilised at 0.17 kPa/d. This study demonstrates that the combined DAF-MF process responded well to substantial changes in raw water quality. In addition, it was suggested that the DAF process must be optimised to avoid excessive membrane fouling.

Case Report: Concurrent Clonidine Abuse and Opioid Use Disorder.

Clonidine is a centrally acting alpha-2 selective adrenergic receptor agonist used to treat hypertension and to control or prevent withdrawal in patients with opioid and alcohol use disorders. Case reports describe abuse of clonidine alone or in combination with benzodiazepines, methadone, codeine, or heroin. Clonidine reportedly boosts and extends the opioid-related euphoria and reduces the amount of psychoactive drug needed. In this case report, we describe clonidine abuse and withdrawal management in an elderly patient with concurrent opioid use disorder. The usage of clonidine in the treatment of opioid detoxification remains controversial. Clonidine abuse is underestimated and requires more attention among health-care providers who concurrently prescribe clonidine and opioids. With the opioid epidemic becoming increasingly prevalent, physicians and other health-care providers must be vigilant in their opioid prescribing as well as concurrent prescribing of other psychoactive pharmacologic agents.

Patient-reported outcomes in light of supportive medications in treatment-naïve lung cancer patients.

PURPOSE: The impact of supportive medications on patient-reported outcomes (PROs) has not been systematically evaluated. We describe the supportive medications used by treatment-naïve lung cancer patients and assess their association with PROs from MD Anderson Symptom Inventory (MDASI). METHODS: Treatment-naïve lung cancer patients who completed PROs from MDASI at the initial visit to MD Anderson Cancer Center were included. Medications from the initial visit were abstracted from the electronic medical records system and categorized into therapeutic classes based on U.S. Pharmacopeia v7.0. A chi-square or Mann-Whitney U test was conducted as appropriate. RESULTS: Among 459 patients, ~ 50% took any analgesics and 25% were on opioids. One-third of patients with moderate-severe pain were not on any analgesics. Patients taking opioids had significantly worse median pain scores (6 vs. 0) compared with those not taking any analgesics (p < 0.0001). Higher proportion of patients with moderate-severe pain took opioids compared with those with mild pain (52% vs. 16%, p < 0.0001). Patients on opioids also reported significantly worse scores for five other cancer-specific core symptoms and all six symptoms rating interference with daily life. Only 15% of patients with higher composite score for depression-related symptoms were on antidepressants. However, patients taking antidepressants did not significantly differ in any individual MDASI symptom scores compared with those not on antidepressants (p = 0.4858). CONCLUSIONS: Our results suggest a need for better screening for pain and depression and optimization of pain management in treatment-naïve lung cancer patients since their poor functional status may result in suboptimal cancer therapy.

Three-dimensional cone beam computed tomographic image reorientation using soft tissues as reference for facial asymmetry diagnosis.

OBJECTIVE: To investigate discrepancies in results of facial asymmetry analysis using different cone beam computed tomography (CBCT) image reorientation methods and the effectiveness of soft tissue as a reorientation reference for analysis of facial asymmetry. MATERIALS AND METHODS: An asymmetric group of 30 patients with 4 mm or more of chin point (menton [Me]) deviation and a symmetric group of 30 patients with less than 4 mm of deviation of Me were chosen as study subjects. Three orientation methods were used to calculate and compare Me deviation values of the 60 subjects. Two methods used only skeletal landmarks for reference, and one method included the soft tissue landmarks around the eye. Preferences of an expert group for the facial midline as determined by each reorientation method were also examined. RESULTS: The examinations showed significant discrepancies in Me deviation values between the three reorientation methods. The expert group showed the greatest preference for the facial midline reorientation method that incorporated soft tissue landmarks of the eye. CONCLUSIONS: These study findings suggest that the inclusion of soft tissue landmarks, especially those around the eyes, is effective for three-dimensional CBCT image reorientation for facial asymmetry analysis.

Gate insulator effects on the electrical performance of ZnO thin film transistor on a polyethersulphone substrate.

Low temperature processing for fabrication of transistor backplane is a cost effective solution while fabrication on a flexible substrate offers a new opportunity in display business. Combination of both merits is evaluated in this investigation. In this study, the ZnO thin film transistor on a flexible Polyethersulphone (PES) substrate is fabricated using RF magnetron sputtering. Since the selection and design of compatible gate insulator is another important issue to improve the electrical properties of ZnO TFT, we have evaluated three gate insulator candidates; SiO2, SiNx and SiO2/SiNx. The SiO2 passivation on both sides of PES substrate prior to the deposition of ZnO layer was effective to enhance the mechanical and thermal stability. Among the fabricated devices, ZnO TFT employing SiNx/SiO2 stacked gate exhibited the best performance. The device parameters of interest are extracted and the on/off current ratio, field effect mobility, threshold voltage and subthreshold swing are 10(7), 22 cm2/Vs, 1.7 V and 0.4 V/decade, respectively.

Chair alarm for patient fall prevention based on gesture recognition and interactivity.

The Gesture Recognition Interactive Technology (GRiT) Chair Alarm aims to prevent patient falls from chairs and wheelchairs by recognizing the gesture of a patient attempting to stand. Patient falls are one of the greatest causes of injury in hospitals. Current chair and bed exit alarm systems are inadequate because of insufficient notification, high false-alarm rate, and long trigger delays. The GRiT chair alarm uses an array of capacitive proximity sensors and pressure sensors to create a map of the patient's sitting position, which is then processed using gesture recognition algorithms to determine when a patient is attempting to stand and to alarm the care providers. This system also uses a range of voice and light feedback to encourage the patient to remain seated and/or to make use of the system's integrated nurse-call function. This system can be seamlessly integrated into existing hospital WiFi networks to send notifications and approximate patient location through existing nurse call systems.