Algorithm For Early Detection Research Articles

Machine learning algorithms for early sepsis detection in the emergency department: A retrospective study

BackgroundEarly recognition and treatment of sepsis are crucial for improving patient outcomes. However, the diagnosis of sepsis remains challenging because of vague clinical presentations. ObjectivesWe aim to developed novel sepsis screening tools with machine learning models and compared their performance with traditional methods. MethodsWe used machine learning algorithms to develop models for early risk prediction of sepsis based on retrospective single-center electronic health record data from adult patients who presented to the emergency department (ED) from June 2018 through May 2020. Available triage data including vital signs, baseline characteristics, and chief complaints served as predictors. In our study, 80% and 20% of the data were randomly split into training and testing sets, respectively. Derived from the training set, we built the models based on four machine learning algorithms: logistic regression, gradient boosting, random forest, and neural network. Our primary outcome was the model performance that predicted the final diagnosis of sepsis determined by the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and predictive performance compared with those of the reference models (quick sequential organ failure assessment [qSOFA], modified early warning score [MEWS], and systemic inflammatory response syndrome [SIRS]) using the testing dataset. ResultsIn total, 133,707 ED visits were analyzed. All machine learning models outperformed the reference models by achieving a higher AUROC (e.g., AUROC 0.931 [95% CI 0.921–0.944] in our best model (random forest algorithm) vs 0.635 [95% CI 0.613–0.660] in qSOFA, 0.688 [95% CI 0.662–0.715] in MEWS, and 0.814 [95% CI 0.794–0.833] in SIRS). ConclusionThe machine learning models demonstrated superior performance in prediction of sepsis diagnosis among emergency patients compared with that using the traditional screening tools. Further studies are needed to determine whether the models will enhance physicians’ judgments and improve patient outcomes.

A machine learning algorithm for early detection of heel deep tissue injuries based on a daily history of sub-epidermal moisture measurements.

Sub‐epidermal moisture is an established biophysical marker of pressure ulcer formation based on biocapacitance changes in affected soft tissues, which has been shown to facilitate early detection of these injuries. Artificial intelligence shows great promise in wound prevention and care, including in automated analyses of quantitative measures of tissue health such as sub‐epidermal moisture readings acquired over time for effective, patient‐specific, and anatomical‐site‐specific pressure ulcer prophylaxis. Here, we developed a novel machine learning algorithm for early detection of heel deep tissue injuries, which was trained using a database comprising six consecutive daily sub‐epidermal moisture measurements recorded from 173 patients in acute and post‐acute care settings. This algorithm was able to achieve strong predictive power in forecasting heel deep tissue injury events the next day, with sensitivity and specificity of 77% and 80%, respectively, revealing the clinical potential of artificial intelligence‐powered technology for hospital‐acquired pressure ulcer prevention. The current work forms the scientific basis for clinical implementation of machine learning algorithms that provide effective, early, and anatomy‐specific preventive interventions to minimise the occurrence of hospital‐acquired pressure ulcers based on routine tissue health status measurements.

Optimization of VGG16 utilizing the Arithmetic Optimization Algorithm for early detection of Alzheimer’s disease

Early detection and prevention of Alzheimer’s disease (AD) is an important and challenging task. Determining a precise and accurate diagnosis of Alzheimer’s disease in its early stages is the most significant challenge. As a result, various research for the early detection of Alzheimer’s disease was conducted. However, these techniques have a number of drawbacks, including higher computational costs, failure to incorporate data from multiple modalities, performance degradation due to data distributions between training and testing data, inability to record brain affected regions, longer processing time, etc. To tackle these issues, we proposed Optimized VGG-16 architecture using Arithmetic Optimization Algorithm (Optimized VGG-16 using AOA) for AD classification. Three major components are involved in this study such as pre-processing, segmentation, and classification. The CAT12 toolkit is used to process the format of T1-weighted MRI images during pre-processing. The image enhancement techniques normalize the uneven light distribution in which the linear contrast stretching enhances the image contrast level. Finally, an Optimized VGG-16 using AOA effectively classifies the AD classes such as normal, mild dementia (severe cognitive decline), and late dementia (very severe cognitive decline) classes. The dataset images are chosen from Alzheimer’s disease Neuroimaging Initiative (ADNI), the Open Access Series of Imaging Studies (OASIS) dataset, and Single Individual volunteer for Multiple Observations across Networks (SIMON) databases. The experimental investigations provided superior classification performances than other existing methods.

Transforming Healthcare with Artificial Intelligence: Innovations, Applications, and Future Challenges

Background: The integration of artificial intelligence (AI) in healthcare has significantly transformed clinical practices, offering substantial improvements in diagnosis, treatment planning, and patient outcome predictions. AI technologies, including artificial neural networks, fuzzy expert systems, and hybrid intelligent systems, are advancing the field of augmented medicine by combining AI with traditional healthcare practices. Methods: This study reviews the diverse applications of AI in healthcare, focusing on its impact on clinical procedures, disease detection, and healthcare management. The analysis covers the use of AI-driven tools such as surgical navigation systems, augmented reality for pain management, and machine learning algorithms for early disease detection and clinical documentation. Results: AI technologies like AccuVein and augmented reality headsets have enhanced clinical procedures such as intravenous placements and surgical interventions. Advances in machine learning, particularly neural networks and deep learning, have improved the detection of complex patterns in imaging data, facilitating early diagnosis of diseases like cancer and pneumonia. Natural language processing (NLP) has enhanced the analysis and classification of clinical documentation, while robotic process automation (RPA) has optimized administrative tasks. AI's role in managing infectious diseases, particularly during the COVID-19 pandemic, has been critical, demonstrating its potential in screening, diagnosis, and treatment surveillance. AI applications in oncology and laboratory medicine have also shown increased accuracy and efficiency in disease diagnosis and patient care. Conclusion: AI is revolutionizing healthcare by enhancing diagnostic accuracy, treatment efficacy, and patient care quality. Despite its transformative potential, challenges such as legal accountability and data bias must be addressed for successful integration into healthcare systems. Continued research and innovation in AI applications are essential to maximizing its benefits while minimizing associated risks.

Enhanced crow search algorithm for early detection of Parkinson's disease in physically challenged patients

Enhanced crow search algorithm for early detection of Parkinson's disease in physically challenged patients

Enhanced crow search algorithm for early detection of Parkinson's disease in physically challenged patients

Enhanced crow search algorithm for early detection of Parkinson's disease in physically challenged patients

New diagnostic algorithm for early detection of gastric cancer: 4D protocol

<h3></h3> Рак желудка по-прежнему остается актуальной проблемой здравоохранения, занимает шестое место среди самых распространенных онкологических заболеваний и четвертое — среди причин смерти от них в мире. Ежегодно диагностируется более 1 млн новых случаев рака желудка. Результаты лечения рака желудка и прогноз зависят от стадии заболевания: 5-летняя выживаемость составляет 25—30%, однако этот показатель у больных ранним раком желудка (T1) после оперативного лечения достигает 95%. К сожалению, более 40% всех случаев рака желудка по-прежнему диагностируется на IV стадии. Ранний рак желудка составляет лишь 5—10% от общего числа больных с этой патологией в Европе, США и России, в то время как в Японии достигает 50%. Из всех методов диагностики эндоскопическое исследование желудка (гастроскопия) является наиболее эффективным в диагностике предопухолевой патологии и рака на ранних стадиях развития опухоли. В связи с этим вопросы совершенствования эндоскопических методик диагностики рака этой локализации, повышения качества эндоскопического исследования остаются по-прежнему в центре внимания широкого круга онкологов, гастроэнтерологов, эндоскопистов. Современная гастроскопия, проводимая с целью диагностики предраковых изменений и ранних форм рака, предполагает использование всех современных технологий, позволяющих выявлять патологические образования даже самых небольших размеров: увеличительной эндоскопии, хромоскопии и узкоспектральной эндоскопии. Наиболее эффективное и рациональное сочетание этих диагностических методик с целью визуализации очаговых структурных изменений слизистой оболочки желудка возможно при использовании современного диагностического алгоритма в повседневной клинической практике. Совершенствование последнего путем включения в него новых диагностических технологий, позволяющих провести эффективную дифференциальную диагностику неопластических изменений, создаст условия для повышения эффективности диагностики предраковых изменений слизистой оболочки и раннего рака желудка у пациентов, подлежащих скринингу.

Standard Clinical Screening Tests, Sural Radial Amplitude Ratio and F Wave Latency Compared to Conventional Nerve Conduction Studies in the Assessment of Sensorimotor Polyneuropathy in Patients with Type 2 Diabetes Mellitus.

Background and Aims:The measuring tools used for assessment of neuropathy include various questionnaires, monofilament testing, Biothesiometry and the gold standard test, nerve conduction studies (NCS). This study aims to evaluate the diagnostic accuracies of Michigan Neuropathy Screening Instrument (MNSI), Biothesiometry, Semmes Weinstein Monofilament (SWMF), Sural Radial Amplitude Ratio (SRAR) and minimal F wave latency as compared to conventional NCS and arrive at a simple diagnostic algorithm for early detection of Diabetic Peripheral Neuropathy (DPN).Methods:In a cross-sectional observational study on 48 Type 2 diabetes mellitus patients, MNSI, Biothesiometry, SWMF and NCS including F waves and SRAR were done and diagnostic accuracies (sensitivity, specificity, positive and negative predictive values) calculated taking NCS as gold standard.Results:MNSI, Biothesiometry, SWMF, SRAR and minimal F wave latency had a sensitivity of 64.3%, 78.6%, 14.3%, 100% and 78.6% and specificity of 67.7%, 52.9%, 94.1%, 23.53% and 76.47% respectively, with reference to NCS. Based on combined sensitivities and specificities, we arrived at a simple algorithm for early diagnosis of DPN, which showed that DPN could either be diagnosed or ruled out in 75% of the patients by a combination of the Biothesiometry, SRAR and left lower limb minimal F wave latency results.Conclusions:In the setting of an outpatient, multidisciplinary diabetic clinic, simple tests such as questionnaires, monofilament testing and biosthesiometer could be performed with greater ease while considering NCS as the gold standard. This algorithm, combining Biothesiometry, SRAR and left lower limb minimal F wave latency would be less time consuming and help in early diagnosis of DPN.

Screening Tests for Early Diagnosis of Multiple Myeloma and Related Plasma Cell Disorders

Monoclonal gammopathy (MG) encompasses a diverse group of disorders characterized by the secretion of monoclonal immunoglobulins or their light-chain components. The incidence of multiple myeloma (MM) in South Korea is rapidly increasing, and it is important to be aware of its initial clinical presentations and the most efficient laboratory algorithms for early detection. Serum protein electrophoresis (SPE) and urine protein electrophoresis (UPE) are the primary screening tests for patients with clinically suspected MM or amyloid light-chain amyloidosis; these tests are reimbursed in South Korea. We reviewed clinical studies that applied national and international guidelines to evaluate test panels for early detection of MGs, including MM. The serum free light chain (sFLC) with SPE panel is recommended for the initial work up for diagnosis of MGs. In the case of a normal SPE, sFLC should be measured subsequently, so as not to miss the presence of M-protein. Use of this screening panel avoids medical expenses related to delayed diagnosis. Guidelines and recommendations suggest that no single method (SPE, serum immunofixation electrophoresis, sFLC, or UPE) should be used to exclude a diagnosis of MM. We believe that a screening test panel comprising SPE plus sFLC will increase the rate of early and accurate diagnosis of MM and related disorders.

Heat Load-Induced Changes in Lying Behavior and Lying Cubicle Occupancy of Lactating Dairy Cows in a Naturally Ventilated Barn

Abstract Dairy cows show a high sensitivity to changes in barn climate, which can result in physiological and ethological responses because of the homeostatic mechanisms to regulate the body temperature under heat load. The objective of this study was to analyze the lying behavior and occupancy of lying cubicles of lactating high-yielding Holstein-Friesian cows throughout the day during three summer months and three winter months. The study was conducted in summer 2016 and in winter 2016/17 in a naturally ventilated barn in Brandenburg, Germany. The determined temperaturehumidity index (THI) of the barn was calculated using the measured ambient temperature and relative humidity at eight locations inside the barn. The THI was used to define the heat load the cows were exposed to. The activity of the cows was measured with accelerometers, and a video recording was made to analyze the occupancy of the three rows of lying cubicles. The results indicated that increasing heat load led to a decrease in lying time; therefore, the daily lying time differed between summer and winter months. In addition, there were different patterns of lying behavior during the course of the day, depending on the season. A sharp decline in lying time could be observed especially in the afternoon hours during the summer. The occupancy of lying cubicles was also influenced by the heat load. The data could be helpful to enable evaluation with algorithms for early detection of heat load.

A Power Tracking Algorithm for Early Detection of Centrifugal Flow Pump Thrombosis.

Logfiles from the HeartWare HVAD System provide operational pump trend data to aid in patient management. Pump thrombosis is commonly associated with increases in the logfile power that may precede the clinical presentation. A Power Tracking algorithm was developed to detect significant deviations in pump power that may be associated with pump thrombus (PT). The Power Tracking algorithm was applied retrospectively to logfiles captured in the ENDURANCE, ENDURANCE Supplemental, and LATERAL clinical trials. From a combined dataset of 896 patients, available logfiles with suspected PT (n = 70 events in 60 patients) and available logfiles from patients without adverse events (AEs) (n = 106 patients, consisting of 27.4 patient-years of monitoring) were organized into two cohorts. The Power Tracking algorithm detected PT cases on or before the recorded AE date with a sensitivity of 85.7%, with detection occurring an average of 3.9 days before clinical presentation. The algorithm averaged one false alarm for every 6.85 patient-years of monitoring from logfiles without AEs. The favorable performance of the Power Tracking algorithm may enable earlier detection of pump thrombosis and allow early medical management versus surgical intervention.

STUDY ON EFFECTS OF RGPO TO RANGE TRACKING SYSTEM OF TARGETING RADAR

The article informs reader about the research results of characteristic features and affecting factors of RGPO, which appear in the range tracking system of Doppler mono-pulse targeting radars. Based on the research results, necessary requirements for security of the range tracking system are formulated. The effective anti-RGPO methods are proposed.&#x0D; Effective control of RGPO requires special method, including in the development special mathematical algorithms for early detection, analysis and recognition escape laws, as well as the method error compensation introduced by RGPO in the tracking system.&#x0D; Alternative algorithm for early detection RGPO and their recognition is proposed, which makes it possible to construct a structure Tracking System with new detector of RGPO.

Plant drought impact detection using ultra-high spatial resolution hyperspectral images and machine learning

Early drought stress detection is crucial for restoring productivity, ensuring recovery, and providing vital information for mortality prevention. Hyperspectral remote sensing which is sensitive to subtle changes in leaf constituents and canopy structure is widely used to study how drought stress manifests and evolves over time. Drought impacts are complicated, affecting multiple leaf properties at different stages of severity. Spectral indices are widely used for stress monitoring, given their sensitivity to one or more plant biochemical, biophysical, and structural properties. However, appropriate spectral bands are not always apparent for early drought detection and stress progression monitoring since drought response varies by species, from tolerators to competitors. A tolerator can produce secondary compounds to enhance leaf survival while a competitor allows stressed leaves to perish so that new growth can occur. Also, spectral indices typically utilize a few bands and therefore do not exploit information in other spectral regions that are also influenced by drought. With the development of machine learning and increased computational power, the use of full spectra and derivative spectra becomes possible, holding great potential to offer new insights into the drought impacts. In this study, full spectra and derivative spectra were integrated with advanced machine learning algorithms for early drought detection and classification. The results from these models were compared to those from traditional spectral index-based methods. The models were tested using drought-stressed smooth brome (Bromus inermis) grass grown in a chamber with 0% and 25% field precipitation, using close-range hyperspectral images. The results showed that the use of derivative spectra and the deep learning model allowed accurate drought detection with an overall accuracy of up to 97.5% and 100% while spectral indices and spectral index models either failed to detect early stress or could detect early stress but with much lower accuracy. This study demonstrated the usefulness of integrating deep learning and hyperspectral data in plant stress research for stress impact assessment, recovery, and mitigation.

Detection of accelerated tool wear in turning

There are many algorithms for the identification of gradual tool wear (GTW) which usually counts in minutes and the detection of catastrophic tool failure (CTF) which counts in milliseconds. However, the tool may lose its cutting ability by accelerated tool wear (ATW), which may last several seconds and cannot be sensed either by GTW or CTF detection algorithms. The paper presents an innovative algorithm for early detection of ATW and CTF. It consists in comparing the waveforms of the cutting force sensor signal in hierarchical time windows. The compared waveforms are independent of the absolute value of the signal. This allows the detection of ATWs of different intensity and duration. Tests proved that successful detection of ATW allows for prevention of CTF.

Multiscale Parallel Algorithm for Early Detection of Tomato Gray Mold in a Complex Natural Environment.

Plant disease detection technology is an important part of the intelligent agricultural Internet of Things monitoring system. The real natural environment requires the plant disease detection system to have extremely high real time detection and accuracy. The lightweight network MobileNetv2-YOLOv3 model can meet the real-time detection, but the accuracy is not enough to meet the actual needs. This study proposed a multiscale parallel algorithm MP-YOLOv3 based on the MobileNetv2-YOLOv3 model. The proposed method put forward a multiscale feature fusion method, and an efficient channel attention mechanism was introduced into the detection layer of the network to achieve feature enhancement. The parallel detection algorithm was used to effectively improve the detection performance of multiscale tomato gray mold lesions while ensuring the real-time performance of the algorithm. The experimental results show that the proposed algorithm can accurately and real-time detect multiscale tomato gray mold lesions in a real natural environment. The F1 score and the average precision reached 95.6 and 93.4% on the self-built tomato gray mold detection dataset. The model size was only 16.9 MB, and the detection time of each image was 0.022 s.

Early detection of bark beetle infestation in Norway spruce forests of Central Europe using Sentinel-2

In the past decade, massive outbreaks of bark beetles (Ips spp.) have caused large-scale decline of coniferous-dominated, prevailingly managed forests of Central Europe. Timely detection of newly infested trees is important for minimizing economic losses and effectively planning forest management activities to stop or at least slow outbreaks. With the advancement of Copernicus services, a pair of Sentinel-2 satellites provides a unique remote sensing data source of multi-temporal observations in high spatial resolution on the scale of individual forest stands (although not allowing for individual tree detection). This study investigates the potential for using seasonal trajectories of Sentinel-2 bands and selected vegetation indices in early detection of bark beetle infestation (so–called green-attack stage detection) in Norway spruce monoculture forests in the Czech Republic. Spectral trajectories of nine bands and six vegetation indices were constructed for the 2018 vegetation season using 14 satellite observations from April to November to distinguish four infestation classes. We used a random forest algorithm to classify healthy (i.e., stands not infested) and infested trees with various trajectories of decay. The seasonal trajectories of vegetation indices separated the infestation classes better than did the individual bands. Among the most promising vegetation indices we identified the tasselled cap wetness (TCW) component and normalized difference index constructed from near and shortwave infrared bands. Analysing the inter-annual change of the indices was more promising for early detection than is single-date classification. It achieved 96% classification accuracy on day of year 291 for the tested data set.The algorithm for early detection of tree infestation based on the assessment of seasonal changes in TCW was applied on the time series of Sentinel-2 observations from 2019 and its outputs were verified using field observations of forest conditions conducted on 80 spruce forest plots (located in spruce monoculture stands). The overall accuracy of 78% was achieved for the separation of healthy and green-attack classes. Our study highlights the great potential of multi-temporal remote sensing and the use of shortwave infrared wavelengths for early detection of spruce forest decline caused by bark beetle infestation.

Early cancer detection from genome-wide cell-free DNA fragmentation via shuffled frog leaping algorithm and support vector machine.

Early cancer detection is significant for patient mortality rate reduction. Although machine learning has been widely employed in that context, there are still deficiencies. In this work, we studied different machine learning algorithms for early cancer detection and proposed an Adaptive Support Vector Machine (ASVM) method by synergizing Shuffled Frog Leaping Algorithm and Support Vector Machine (SVM) in this study. Since ASVM regulates SVM for parameter adaption based on data characteristics, the experimental results reflected the robust generalization capability of ASVM on different datasets under different settings; for instance, ASVM can enhance the sensitivity by over 10% for early cancer detection compared with SVM. Besides, our proposed ASVM outperformed Grid Search + SVM and Random Search + SVM by significant margins in terms of the area under the ROC curve (AUC) (0.938 versus 0.922 versus 0.921). The proposed algorithm and dataset are available at https://github.com/ElaineLIU-920/ASVM-for-Early-Cancer-Detection. Supplementary data are available at Bioinformatics online.

DEDA: An algorithm for early detection of topology attacks in the internet of things

The internet of things (IoT) is used in domestic, industrial as well as mission-critical systems including homes, transports, power plants, industrial manufacturing and health-care applications. Security of data generated by such systems and IoT systems itself is very critical in such applications. Early detection of any attack targeting IoT system is necessary to minimize the damage. This paper reviews security attack detection methods for IoT Infrastructure presented in the state-of-the-art. One of the major entry points for attacks in IoT system is topology exploitation. This paper proposes a distributed algorithm for early detection of such attacks with the help of predictive descriptor tables. This paper also presents feature selection from topology control packet fields. The performance of the proposed algorithm is evaluated using an extensive simulation carried out in OMNeT++. Performance parameter includes accuracy and time required for detection. Simulation results presented in this paper show that the proposed algorithm is effective in detecting attacks ahead in time.

Epidemiological and pharmacoeconomic aspects of HIV infection in military

The goals of the research are to determine the latent morbidity and financial losses of the Russian Ministry of Defense due to the dismissal of service members with HIV infection at symptomatic stages; to establish the supposed clinical and pharmacoeconomic effectiveness of the early detection and treatment of service members.The materials compile 379 case histories of patients of the only specialized department for HIV-positive military personnel in the Armed Forces for 2017–2019, as well as various reporting forms. The study uses the following methods for the analysis: Markov chains are employed to calculate QALYs (Quality Adjusted Life Years), “cost-effectiveness”, and “willingness to pay”. The research compares two competitive models: “Current Situation” - without the introduction of algorithms for early detection and treatment of HIV infection, and “Prognosis” - with the introduction of these algorithms.The following results are obtained: implementing a mandatory triennial screening for HIV-antibodies among military personnel allows to raise the detection of HIV-positive military personnel in the early stages of the disease by 55%. At the same time, early administered antiretroviral therapy prevents the progression of HIV infection, while the number of military personnel who have reached symptomatic stages of HIV infection and the number of deaths decreases by 35,6% and 80,2% (six-fold), respectively. The economic effectiveness of these measures includes a reduced cost of treatment and an increased length of military service. The sensitivity analysis of the method shows that the proposed measures remain pharmacoeconomically highly effective with a wide variability of the initial data.Conclusion. The results of the study demonstrate that a significant improvement of the epidemiological situation in the field of HIV infection in the Armed Forces is achievable at very low economic costs. With the full implementation of the proposed measures, the cohort of military personnel of the Armed Forces is able to achieve the World Health Organization’s “90-90-90”criteria of in 3–5 years.

EADA: An Algorithm for Early Detection of Attacks on IoT Resources

IoT applications are becoming widespread in monitoring and managing critical infrastructure. Many attacks have been demonstrated in the state-of-the-art on IoT resources. These attacks make use of vulnerabilities present in various connected systems and the Internet of Things (IoT). The state-of-the-art presents many approaches to detect and mitigate such attacks on IoT resources. The early attack detection mechanism is essential to prevent damage to the IoT system and human. This paper presents an algorithm for early detection of attacks on IoT resources through use of predictive descriptor tables. Effectiveness of the proposed algorithm is evaluated through experimental setup built using Google cloud platform. Experimental results show that the proposed algorithm is efficient in the detection of attacks in real-time.