Windows PC Research Articles

Leveraging code-free deep learning for pill recognition in clinical settings: A multicenter, real-world study of performance across multiple platforms

BackgroundPreventable patient harm, particularly medication errors, represent significant challenges in healthcare settings. Dispensing the wrong medication is often associated with mix-up of lookalike and soundalike drugs in high workload environments. Replacing manual dispensing with automated unit dose and medication dispensing systems to reduce medication errors is not always feasible in clinical facilities experiencing high patient turn-around or frequent dose changes. Artificial intelligence (AI) based pill recognition tools and smartphone applications could potentially aid healthcare workers in identifying pills in situations where more advanced dispensing systems are not implemented. ObjectiveMost of the published research on pill recognition focuses on theoretical aspects of model development using traditional coding and deep learning methods. The use of code-free deep learning (CFDL) as a practical alternative for accessible model development, and implementation of such models in tools intended to aid decision making in clinical settings, remains largely unexplored. In this study, we sought to address this gap in existing literature by investigating whether CFDL is a viable approach for developing pill recognition models using a custom dataset, followed by a thorough evaluation of the model across various deployment scenarios, and in multicenter clinical settings. Furthermore, we aimed to highlight challenges and propose solutions to achieve optimal performance and real-world applicability of pill recognition models, including when deployed on smartphone applications. MethodsA pill recognition model was developed utilizing Microsoft Azure Custom Vision platform and a large custom training dataset of 26,880 images captured from the top 30 most dispensed solid oral dosage forms (SODFs) at the three participating hospitals. A comprehensive internal and external testing strategy was devised, model's performance was investigated through the online API, and offline using exported TensorFlow Lite model running on a Windows PC and on Android, using a tailor-made testing smartphone application. Additionally, model's calibration, degree of reliance on color features and device dependency was thoroughly evaluated. Real-world performance was assessed using images captured by hospital pharmacists at three participating clinical centers. ResultsThe pill recognition model showed high performance in Microsoft Azure Custom Vision platform with 98.7 % precision, 95.1 % recall, and 98.2 % mean average precision (mAP), with thresholds set to 50 %. During internal testing utilizing the online API, the model reached 93.7 % precision, 88.96 % recall, 90.81 % F1-score and 87.35 % mAP. Testing the offline TensorFlow Lite model on Windows PC showed a slight performance reduction, with 91.16 % precision, 83.82 % recall, 86.18 % F1-score and 82.55 % mAP. Performance of the model running offline on the Android application was further reduced to 86.50 % precision, 75.00 % recall, 77.83 % F1-score and 69.24 % mAP. During external clinical testing through the online API an overall precision of 83.10 %, recall of 71.39 %, and F1-score of 75.76 % was achieved. ConclusionOur study demonstrates that using a CFDL approach is a feasible and cost-effective method for developing AI-based pill recognition systems. Despite the limitations encountered, our model performed well, particularly when accessed through the online API. The use of CFDL facilitates interdisciplinary collaboration, resulting in human-centered AI models with enhanced real-world applicability. We suggest that rather than striving to build a universally applicable pill recognition system, models should be tailored to the medications in a regional formulary or needs of a specific clinic, which can in turn lead to improved performance in real-world deployment in these locations. Parallel to focusing on model development, it is crucial to employ a human centered approach by training the end users on how to properly interact with the AI based system to maximize benefits. Future research is needed on refining pill recognition models for broader adaptability. This includes investigating image pre-processing and optimization techniques to enhance offline performance and operation on handheld devices. Moreover, future studies should explore methods to overcome limitations of CFDL development to enhance the robustness of models and reduce overfitting. Collaborative efforts between researchers in this domain and sharing of best practices are vital to improve pill recognition systems, ultimately enhancing patient safety and healthcare outcomes.

An expandable voice user interface as lab assistant based on an improved version of Google’s speech recognition

Voice assistants are potentially helpful when working in a scientific laboratory. A big challenge is the extremely specific use of language in every laboratory. As with any voice assistant, another concern is data security. Here, we present Rainbow—an open source voice user interface (VUI) for scientific laboratories, that is adaptable to any Windows PC with Internet access. We used Google Translate Site (GTS) as a voice input and output system to ensure communication to the user. The scripting language AutoIt controls GTS, executes all actions and builds the VUI. Rainbow performs tasks from three different areas—general Microsoft Windows tasks, lab-specific tasks, and device-specific tasks. We achieved significantly higher speech recognition accuracy with our VUI than with GTS alone (91.3% versus 85.1%). Because of Rainbow's architecture, it is possible to improve the voice assistant in terms of functionality and accuracy, allowing each laboratory to optimize its own Rainbow system in a user-friendly way. In a test setup, this led to a speech recognition accuracy of 98.6%. Taken together, Rainbow provides an opportunity for every scientist to implement highly specific scientific terms and tasks to this open source voice assistant system in a very user-friendly way.

Implementation and Evaluation of a Federated Learning Framework on Raspberry PI Platforms for IoT 6G Applications

With the advent of 6G technology, the proliferation of interconnected devices necessitates a robust, fully connected intelligence network. Federated Learning (FL) stands as a key distributed learning technique, showing promise in recent advancements. However, the integration of novel Internet of Things (IoT) applications and virtualization technologies has introduced diverse and heterogeneous devices into wireless networks. This diversity encompasses variations in computation, communication, storage resources, training data, and communication modes among connected nodes. In this context, our study presents a pivotal contribution by analyzing and implementing FL processes tailored for 6G standards. Our work defines a practical FL platform, employing Raspberry Pi devices and virtual machines as client nodes, with a Windows PC serving as a parameter server. We tackle the image classification challenge, implementing the FL model via PyTorch, augmented by the specialized FL library, Flower. Notably, our analysis delves into the impact of computational resources, data availability, and heating issues across heterogeneous device sets. Additionally, we address knowledge transfer and employ pre-trained networks in our FL performance evaluation. This research underscores the indispensable role of artificial intelligence in IoT scenarios within the 6G landscape, providing a comprehensive framework for FL implementation across diverse and heterogeneous devices.

Expert knowledge–based system for risk assessment of the occurrence of Amyloodinium ocellatum in semi-intensive fish farms

The implementation of a system to assess the risk of Amyloodinium ocellatum occurrence in rearing ponds in fish farms located in southern Spain is a fundamental aspect to ensure the economic viability of these facilities. For this purpose, a computer program (called Amy) for Windows PCs and an application for mobile devices (AmyAPP), based on the Android operating system, were developed integrating transformation functions and weightings associated with environmental parameters and fish behavioural factors from which it is possible to estimate the level of risk of occurrence of A. ocellatum. The weights for each of the environmental parameters and behavioural factors were estimated from the responses of a panel of experts (the fish farmers) using a Delphi methodology. The results indicate that, under operational validation, Amy/AmyAPP responses were statistically sensitive to the occurrence of A. ocellatum outbreaks in sea bream (Sparus aurata) and sea bass (Dicentrarchus labrax) rearing ponds.

SPARSHA-AR an Android Application for the Visualization of Aero Engine Components’ Assemblies and Sub-Assemblies Using Augmented Reality Environment

Abstract: In this project we are applying the concept of Augmented Reality (AR) to display the Aero engine components’ assemblies and sub assemblies. We are displaying those components in Android Smartphone in more attractive way by enabling the user to interact with those components’ using Android Smartphone and Windows PC. Also, as part of the project we have made it possible to display the information of respective model along with voice read out complete description of model simultaneously.

Interactive Folklore - Re-Modernizing The Culture Through Digital Storytelling

This project focuses on developing an interactive visual novel prototype based on a popular Malaysian folklore, Batu Belah Batu Bertangkup. The aim is to modernize and digitize folklore as an interactive visual novel with branching narratives for teaching children about moral values through the story. This visual novel includes user interaction, core gameplay mechanics, storyline, character design, and backgrounds. The game is developed using Ren’Py for Windows PC as the platform

An insight into gender based variations of autonomic functions among young adults

Since this is a vital developmental phase, studying the health patterns of young adults, particularly cardiovascular health, is critical. Autonomic function testing (AFT) is a crucial non-invasive procedure for determining the state of one's cardiovascular system. There are only a few studies on autonomic responsiveness in young people. To study the autonomic responses to various cardiovascular autonomic function tests among young adult subjects and to look into the gender-based variations.Males made up 114 (50.67 percent) of the total 225 individuals, while females made up the rest. The study's target age group was 18-29 years old. The CANWIN analyser of genesis medical systems was used to investigate autonomic function tests such as resting heart rate, heart rate reaction to slow deep breathing (SDB), Valsalva Ratio (VR), 30:15 ratios response to standing, and blood pressure response to sustained handgrip. CANWIN is a state-of-the-art Cardiac Autonomic Neuropathy (CAN) Analysis System with interpretation for PC Windows*. The data was analysed statistically using STATA software (version 13.1).: Dysautonomia was seen in 150 of the 225 participants (66.66 percent). The number of dysautonomia participants was not statistically different between males and females (P value=0.15).: Anxiety, stress, a poor diet, and a disrupted sleep cycle could all contribute to the high number of dysautonomia patients. There have been a number of reports linking dysautonomia to nutritional deficits. To investigate the causative factors, more longitudinal studies including nutritional status assessments are needed.

DETECTABILITY AND ANATOMICAL CORRELATION OF MIDDLE EAR CHOLESTEATOMA USING NON ECHO PLANAR DIFFUSION-WEIGHTED IMAGING AND HIGH RESOLUTION COMPUTED TOMOGRAPHY

This study was conducted to evaluate the role of non-echoplanar diffusion weighted imaging with apparent decient coefcient values and high resolution computed tomography in detection and anatomical extend of middle ear cholesteatoma, the extent of invasion of cholesteatoma using diffusion weighted imaging and high resolution computed tomography, and ability of MRI non-echoplanar DWI sequence unequivocally conrming the diagnosis of cholesteatoma. A cross sectional study with a t Materials and method: otal of 32 patients was conducted at Chettinad Hospitals and Research Institute's Department of Radiology between January 2020 and July 2021 after being diagnosed with chronic otitis media using otoscope as per inclusion and exclusion crietria . Written informed consent was taken to participate in the study, which was authorised by our institution's ethics committee. All statistical analyses we Results: re carried out on Microsoft Windows PCs using the SPSS, version 17. The descriptive data was represented using numbers and %ages. The data was expressed using the mean and standard deviation. To analyse qualitative data, the chi-square test was used. The screening test was carried out utilising a receiver operating characteristic (ROC) curve developed by the researchers. A p value 0.05 was judged statistically signicant on both sides of the equation. Combining DWI with HRCT is a Conclusion: valuable method for evaluating the soft tissue of the middle ear. It has high sensitivity and specicity for detecting all primary cholesteatomas in their various stages of development

PLEURO-CUTANEOUS WINDOW IN MANAGEMENT OF CHRONIC EMPYEMA- A SINGLE CENTER STUDY

Background: Various treatment options are available for the management of empyema including medical therapy, open and closed drainage, decortication and thoracoplasty. Pleuro-cutaneous window (PC window) is an accepted method of treatment for optimal drainage of chronic empyemas. Open window thoracostomy should be considered in debilitated patients with empyema, where denitive surgery is not indicated and failure to control the disease by tube thoracostomy. Materials and Method: This was a single center study, conducted on 60 patients of chronic empyema at Institute of Respiratory Diseases during January 2016 to December 2020. These patients were initially treated with medical therapy and chest tube drainage but their lungs were not completely expanded. They were not the candidates for denitive surgery at that time because of their medical and lungs condition, hence they were treated by pleura-cutaneous window. Results: Mean age of patients in our study was 35.63 ± 12.14 years and age range was 15 to 60 years. In our study 49 patients were male and 11 were females. Patients with left sided empyema were 34 (56.67%) and with right sided empyema were 26 (43.33%). Most of the patients 44 (73.33%) had chest tube drainage for 3 to 4 months. Average duration of hospital stay was 7 days. Lung expansion occurred in 32 (53.33%) patients. Conclusion: Open drainage by pleuro-cutaneous window expands the entrapped lungs despite the presence of broncho-pleural stulas. Pleuro-cutaneous window also has psychological benets in these patients

학생선수의 조직공정성, 감정고갈, 중도포기의도의 구조적 관계

The purpose of this research is to investigate a structural relation of organizational justice, emotional exhaustion, intention to dropout for student athletes. To achieve this purpose, 348 student athletes studying at Seoul, Gyeonggi Province as of 2022 were selected for analysis. In terms of the sampling method, this study used convenience sampling and it extracted 345 subjects for analysis. Data processing was analyzed and verified with SPSS 23.0 version and Amos 23.0 version for PC Windows. The specific results are following. First, organizational justice of student athletes affects emotional exhaustion negative(-). Second, organizational justice of student athletes affects intention to dropout negative(-). Third, emotional exhaustion of student athletes affects intention to dropout positive(+). In summary, organizational justice of student athletes had an effect on intention to dropout through emotional exhaustion. Therefore, this study provides the fact that organizational justice of student athletes is a major factor in understanding the overall process of intention to dropout. Furthermore, as a way to minimize the intention to dropout of student athletes, it explains that the importance of organizational justice and the alleviation of emotional exhaustion are the main issues.

“Master Teacher” Transformative Learning Strategy in Digital Media: “Ruangguru” in Indonesia

Ruangguru is a digital media learning application created by a startup that understands Generation Z. It provides teaching materials for elementary, junior high, and high school children that can be accessed via Android, iOS, and PC (Windows and macOS), as part of their daily use. This breakthrough was made to maintain school-aged children’s interest and motivation to learn, as Ruangguru carries the tagline: Belajar Jadi Luar Biasa (Learning Becomes Extraordinary) through facilities such as videos, practice exercises, private lessons, as well as tryouts. The novelty in this study lies in the concept of "Master Teacher". Based on the observation and interviews conducted, Ruangguru uses the term "Master Teacher" to refer to the instructors, which raises an understanding of the quality instructors Ruangguru has. Therefore, the researchers intend to reveal the Master Teacher’s learning strategy during the online learning process through digital media. The objectives of this study are as follows: 1) to explore the Master Teacher’s learning strategy while preparing interesting digital media-based applications; 2) to identify Ruangguru’s target audience 3) to explore reasons for using Master Teacher as part of the learning strategy. The research has used case study as a qualitative method to collect data through interviews, observation, and literature studies. The results of this study indicate that: 1) Ruangguru as a learning application provides content for various levels in an interesting way through videos, practice questions, private lessons, tryouts which can be accessed from a smartphone; 2) Ruangguru’s content is adapted from the national curriculum and specifically designed by the best and experienced teachers. It is aimed at Generation Z ranging from elementary, junior high, to high school students; 3) Master Teacher (MT), as Ruangguru’s strong point, serves as a representative in delivering digital media-based transformative learning strategies to Generation Z; 4) Transformative learning strategy is used by Ruangguru Master Teacher (MT) because it suits Generation Z’s characteristics that like Ruangguru’s digital media as a platform to deliver messages. Since the urgency and potential of learning through digital media are considered as an important finding in this research, it is recommended that the results of this study are utilized, disseminated, and developed in learning processes.

Improved Real-Time Traffic Obstacle Detection and Classification Method Applied in Intelligent and Connected Vehicles in Mixed Traffic Environment

Mixed traffic is a common phenomenon in urban environment. For the mixed traffic situation, the detection of traffic obstacles, including motor vehicle, non-motor vehicle, and pedestrian, is an essential task for intelligent and connected vehicles (ICVs). In this paper, an improved YOLO model is proposed for traffic obstacle detection and classification. The YOLO network is used to accurately detect the traffic obstacles, while the Wasserstein distance-based loss is used to improve the misclassification in the detection that may cause serious consequences. A new established dataset containing four types of traffic obstacles including vehicles, bikes, riders, and pedestrians is collected under different time periods and different weather conditions in urban environment in Wuhan, China. Experiments are performed on the established dataset on Windows PC and NVIDIA TX2, respectively. From the experimental results, the improved YOLO model has higher mean average precision than the original YOLO model and can effectively reduce intolerable misclassifications. In addition, the improved YOLOv4-tiny model has a detection speed of 22.5928 fps on NVIDIA TX2, which can basically realize the real-time detection of traffic obstacles.

PEMBUATAN GAME 2D HORROR “BLACK, WHITE, AND …” DENGAN PLATFORM WINDOWS

"Black, White And ..." is a 2D Horror game with a sub-genre psychological horror game developed for Windows PC. This game was made using Unity as the game engine with C# as the programming language. In this game, the player will play as an amnesiac girl who has just awoken. The goal of this game is to search for a way out from the strange black and white world. The player will have to survive from the enemy and traps with the ability to move from the black world to the white world and vice versa. to ward the enemies, the player had to scream. The game has implemented a Voice Recognition feature to differentiate between scream or not scream. The testing has been done by using blackbox testing, alpha testing, and beta testing method. The test results show that most of the respondents experience a sense of fear and tension as well as the intention to play the game again.

Virtual Colour Atlas

AbstractA web application (Virtual Colour Atlas) has been developed that implements seven of the leading color order systems: Colorcurve, Coloroid, DIN 6164, Munsell, OSA‐UCS, NCS, and Swiss Colour Atlas 2541, three color spaces: CIE xyY, CIELAB, and CIELUV, and three colour ranges: BS:5252, RAL 840‐HR, and RAL 841‐GL—allowing the user to view any of the core planes in the color space of each of these systems. The Virtual Colour Atlas is freely available at (www.vcsconsulting.uk) and provides visual interaction with these systems using any modern browser hosted on a device with a sufficiently large physical display including: windows PCs, Macs, iPads, Linux devices, and smart televisions (HD to 8 K). This paper provides a brief background on the systems, describes the Virtual Colour Atlas software and history, and discusses example applications of the Virtual Colour Atlas.

Experience a Faster and More Private Internet in Library and Information Centres with 1.1.1.1 DNS Resolver.

As a densely populated country with a large user base, the library and information centres (LICs) in India always suffer from a lack of internet speed. Indian LICs frequently employ open DNS resolvers to achieve high internet speeds. In the new normal, the cyberattacks on Indian educational institutions have skyrocketed, necessitating the development of a more secure and quicker DNS resolver. To meet this demand, the paper assesses Cloudflare's 1.1.1.1 DNS resolver and demonstrates its suitability for Indian LICs. To examine the demand of Cloudflare’s 1.1.1.1 DNS resolver in providing the most secure and fastest private network, the researchers use six Windows PCs connected with ten different networks with the same ISP provider and calculate the Internet speed Measurement Lab and the nPerf online free internet speed checker in Chrome browser (v. 96.0). The internet speed was manually collected before and after the configuration and further analysed to evaluate the increase/decrease percentage. The study findings reveal that after the configuration of 1.1.1.1 set up, the internet speed of the chrome browser in the targeted PCs increased 36.33% for download and 73.88% for upload. And from the result, we can conclude that Indian LICs can rely on the 1.1.1.1 DNS resolver to experience fast & secure private network connections. There are several studies that compare two or more DNS resolvers in terms of speed and security, but none of them specifically addresses their usability and benefits in LICs. Although the usage of DNS resolvers at the institutional level is not a new concept, there is still a lack of understanding among library professionals in India about the deployment criteria and benefits. As a result of this research, library professionals may be inspired to use DNS resolvers comprehensively to provide lightning-fast services to their users.

Operating outdated EPMA (JXA-8600SX) by popular microcomputers and Windows PC

Here we report on the operation of the JEOL JXA-8600SX wavelength-dispersive electron probe microanalyzer (EPMA) after replacing the outdated and damaged native computers with popular and up-to-date microcomputers (Arduino) and a desktop Windows PC. Stepper motors actuating the sample stage and spectrometers are driven by computerized numerical control (CNC) kits commercialized for hand-made 3D printers, open-source firmware GRBL installed on Arduino, and a common text-based command “G-code” sent from a PC. X-ray pulse signals from the single-channel pulse-height analyzer (SCA) and analog signals from the probe ammeter are also acquired by Arduino. Sending commands from the PC to Arduino and receiving signals from Arduino to the PC are routinized by a homemade program coded by Visual Basic.NET. Quantitative analyses of anhydrous silicates and oxides, applying the Bence-Albee correction, provide reasonable oxide concentrations that are generally within 99-101 wt.% in total and compatible with stoichiometry within a 1%-2% relative deviation. This alternative control system not only extends the use of EPMA hardware regardless of generational changes in computer technology, but also introduces optional analytical techniques not available in the built-in system.

Pixel calculations using Orca or GAUSSIAN for electron density automated within the Oscail package.

Many discussions of the intermolecular interactions in crystal structures concentrate almost exclusively on an analysis of hydrogen bonding. A simple analysis of atom-atom distances is all that is required to detect and analyse hydrogen bonding. However, for typical small-molecule organic crystal structures, hydrogen-bonding interactions are often responsible for less than 50% of the crystal lattice energy. It is more difficult to analyse intermolecular interactions based on van der Waals interactions. The Pixel program can calculate and partition intermolecular energies into Coulombic, polarization, dispersion and repulsion energies, and help put crystal structure discussions onto a rational basis. This Windows PC implementation of Pixel within the Oscail package requires minimal setup and can automatically use GAUSSIAN or Orca for the calculation of electron density.

An open-source, low-cost NMR spectrometer operating in the mT field regime

In recent years, low field and ultra-low field NMR spectrometers have gained interest due to their portability, lower cost, and reduced subject-induced magnetic field inhomogeneities. Here, we describe the design of a low-cost multinuclear NMR spectrometer operating in the ultra-low field regime (ULF), which possesses high spectral resolution and enables arbitrary pulse programming. An inexpensive multifunction input/output (I/O) device is used to handle waveform generation and digitization in the kHz operating range. A home-built radio frequency (RF) mixing circuit is used to down-mix the NMR signals, allowing for the slower sampling rates and lower memory requirements needed to enable minute-long acquisitions using a standard Windows PC. The LabVIEW code, along with a bill of materials for all components used in the spectrometer, is included. As proof of concept, 1H relaxation measurements and the simultaneous detection of 1H with gas phase and dissolved 129Xe frequencies using the described low field NMR spectrometer are demonstrated.

Open source 3D phenotyping of chickpea plant architecture across plant development

BackgroundBeing able to accurately assess the 3D architecture of plant canopies can allow us to better estimate plant productivity and improve our understanding of underlying plant processes. This is especially true if we can monitor these traits across plant development. Photogrammetry techniques, such as structure from motion, have been shown to provide accurate 3D reconstructions of monocot crop species such as wheat and rice, yet there has been little success reconstructing crop species with smaller leaves and more complex branching architectures, such as chickpea.ResultsIn this work, we developed a low-cost 3D scanner and used an open-source data processing pipeline to assess the 3D structure of individual chickpea plants. The imaging system we developed consists of a user programmable turntable and three cameras that automatically captures 120 images of each plant and offloads these to a computer for processing. The capture process takes 5–10 min for each plant and the majority of the reconstruction process on a Windows PC is automated. Plant height and total plant surface area were validated against “ground truth” measurements, producing R2 > 0.99 and a mean absolute percentage error < 10%. We demonstrate the ability to assess several important architectural traits, including canopy volume and projected area, and estimate relative growth rate in commercial chickpea cultivars and lines from local and international breeding collections. Detailed analysis of individual reconstructions also allowed us to investigate partitioning of plant surface area, and by proxy plant biomass.ConclusionsOur results show that it is possible to use low-cost photogrammetry techniques to accurately reconstruct individual chickpea plants, a crop with a complex architecture consisting of many small leaves and a highly branching structure. We hope that our use of open-source software and low-cost hardware will encourage others to use this promising technique for more architecturally complex species.

Development of Smart Mobile Manipulator Controlled by a Single Windows PC Equipped with Real-Time Control Software

Development of Smart Mobile Manipulator Controlled by a Single Windows PC Equipped with Real-Time Control Software