Search

Automatic Number Plate Recognition (LPR) plays an important role in numerous applications and various techniques have been proposed. In this paper we will learn about the automatic license plate detection of the security purpose and documentation of entry of vehicles inside any building. In this technique we are using easy-OCR for detection of the characters and numbers present in the number plate. By using this technique, we can store the number plate details and vehicle entry time in an accurate rate and in a faster way. In this project we are using the camera to take the picture of the number plate and then by using easy OCR we are taking the characters and number values from the image of the number plate and then the data is stored in the data base and can be exported when ever needed.

In this work, a unique calorimetric system was presented for the determination of horseradish peroxidase (HRP) and hydrogen peroxide (H2O2). This HRP-catalyzed system is based on the formation of electrophilic 1,4 –diimine by the oxidation of Paraphenylenediamine (PPD), this intermediate undergoes coupling with the free para position of Alpha naphthol (AN) to generate the stable intense purple-colored radical with λmax 520nm had good linear with peroxidase concentration 0.08-1.32 nm and 0.02-0.66 nm by kinetic and one-time detection method and H2O2 amid 0.6 – 38.7 µM from the kinetic and 0.07- 4.83 µM by fixed time method, respectively. Moreover, the applicability of the proposed PPD-AN system has been successfully used to detect HRP activity in medicinal plants with crude extract as the source of enzyme and the outcomes were compared with the reference method. Overall, the present system can be adopted as a candidate system for the determination of H2O2.

With covid infection having a periodic occurrence and each wave affecting millions of people, there is an urgent need for healthcare providers to develop readily available biological markers to predict the severity and mortality of COVID-19 infection at an early stage of the disease to predict the most optimal management. The Neutrophil – lymphocyte Ratio (NLR) and Platelet-lymphocyte ratio (PLR), are novel inflammatory markers suggested to predict the severity of illness in COVID -19 patients. We evaluated a retrospective cohort of 224 patients with covid 19 illnesses from August 2020 to October 2020. We categorized the patients into 'mild' and 'moderate to severe as per the severity of their illnesses. The optimal cut-off values of the continuous NLR, PLR, and Age were calculated by applying the receiver operating characteristic (ROC) curve analysis. Sensitivity, specificity, and likelihood ratios were calculated from ROC analysis. The results showed that Age, Hb, and NLR are significant predictors of the severity of illness. So we quantified the predictive contribution of those variables by its Z- score and showed that NLR is the strongest predictor followed by Age.ROC curve analysis revealed NLR to be an excellent predictor of disease severity (sensitivity 88.30% and specificity 82.60%) at a cut-off of > 4.24. A cut-off of 165.63 was obtained for PLR, which showed good predictive value for the severity of illness. A NLR > 6 was used as a predictor for mortality due to covid illness with good sensitivity and specificity. Conclusion: NLR and PLR can be used as prognostic markers to identify the severity of covid illness and the probability of mortality in the future in low-resource settings. Given its low cost, easily available, accessible, and reproducible data, it can be used as a useful marker to assist the physician in early referral and treatment of covid 19 infected patients.

Machine learning and deep learning are branches of artificial intelligence consisting of statistical, probabilistic, and optimisation techniques that allow machines to learn from previous observations recorded by humans. These machine learning algorithms, when combined with other technologies, can be used to perform very intuitive yet awkward human-like tasks. Using these algorithms, humans can enable computers to learn about certain things like recognising an object in an image. Prognosis is an important clinical skill, particularly for cancer patients' clinicians, neurooncologists. One of the biggest challenge for AI in prognosis is to verify and validate its models. Unlike diagnosis, the prognosis models are centered on predictive data that usually addresses the patient and not the disease. Prognosis models were developed to aid in the decision-making of patients' treatment.AI can have bigger impact in the health care domain with more healthcare providers using AI to make the first diagnosis and prognosis more accurate and interpretable with the available patient data for better therapy.

In fields like healthcare, where human intelligence is critical, the introduction of new AI-powered applications is becoming increasingly popular. Technologies have reduced expenses, accelerated drug research, and improved wellness outcomes. AI has become increasingly cognizant of its potential to disrupt the business, as seen by growing funding for the sector in recent years from important stakeholders in both healthcare and risk capital. Traditional approaches include human participation and direct interaction with patients, which are now obsolete due to the advancement of technologies such as messaging bots and intelligent virtual assistants. On the other hand, the internet of things (IoT) is making significant contributions to healthcare, and its gadgets may collect complete health data. Machine intelligence collects and analyses data in established protocols in search of possible health-related predictions. The chapter delves into the aspects of combining AI and IoT to improve efficiency in healthcare systems.

Background As the susceptibility pattern of different pathogens varies among different settings, the evaluation of appropriate clinical diagnosis and timely initiation of the empirical antibiotic treatment based on the local susceptibility data is crucial in the management of sepsis. Methods A retrospective study was conducted among adult patients with sepsis at a charitable hospital in Mangaluru. The essential details such as patient demographics, culture specimens, organisms, resistance/susceptibility pattern, laboratory data, empirical therapy and clinical outcomes were collected from the medical records. Descriptive statistics were used in analysing the data. Results A total of 425 patients diagnosed with sepsis during the study period were screened to meet the sample size of 373 positive cultures, among which 367 (91.3%) samples yielded the bacterial isolates, of which 250 (68.1%) and 117 (31.9%) were gram-negative and gram-positive organisms, respectively. The most common gram-negative organisms isolated were K pneumoniae (19.9%), A baumannii (19.6%) and E coli (12.8%); while Coagulase-negative staphylococcus (14.4%) and S aureus (8.4%) were the predominant gram-positive organisms. The isolated pathogens showed a resistance rate of >50% to the most commonly used antibiotics. Conclusion The present study provides information on the prevalence of the most common pathogens and their resistance pattern to different antibiotics, which plays a vital role in the selection and timely initiation of the appropriate empirical antibiotic therapy.

The Internet of Things has observed tremendous growth in the last two decades utilizing wireless sensor technologies and as the number of smart devices and applications increases, the need for data collection and analysis data becomes increasingly difficult. In general sensor nodes are powered through batteries hence it restricts the efficiency of Wireless Senor Network. Data aggregation is a top-most mechanism for redundancy discarding and improvising the network lifetime. It also possesses various advantages which include minimizing the energy consumption, minimizing the network traffic. Furthermore, as sensor nodes are deployed in hostile averments and transmit sensitive information, designed networks are prone to various attacks. As a result, the focus of this research work is on designing and developing the Nearest Neighbor Monitoring (NN-SDA) mechanism to provide secure data aggregation in an efficient and secured manner. Furthermore, the proposed mechanism allows the nearest neighbour node to monitor and detect the malicious node. Further, the NN-SDA-mechanism algorithm is designed by adding the anomaly and detecting the malicious nodes. The NN-SDA mechanism is evaluated in terms of malicious packet identification rate, average energy utilization, and system throughput and the comparative analysis is performed with the existing model.

Electrohysterogram (EHG) is a noninvasive approach for recording the electrical activity of the uterine muscles (myometrium). It is also known as Uterine Electromyogram (Uterine-EMG). Since it is a non-invasive alternative, it is a safe and painless procedure to monitor uterine activity. In this paper, a wavelet transformation technique was used to preprocess the raw EHG signal, to remove artifacts present in it. A secondary time series was computed from the peak intervals of the EHG signals. This was followed by the extraction of the non-linear features from this time series. The classification of labor and non-labor signals was performed with a KNN algorithm. An accuracy of 90.32% was achieved with the KNN model. With supervised learning models, labor contractions were correctly identified in women who had just entered their third trimester but had not crossed the 37th week mark. With this knowledge, potential cases of preterm births could be identified. This knowledge could assist doctors in premeditating the prevention of still-births.

A middle-aged woman with diabetes presented with left-sided facial pain, complete ptosis and fever of short duration. On presentation, she had hyperglycaemia without ketosis. There was total ophthalmoplegia of the left eye with a visual acuity of 6/36. She incidentally tested positive for COVID-19. CT paranasal sinus and MRI brain revealed left-sided pansinusitis with acute infarct in the left parieto-occipital region without angioinvasion. An emergency functional endoscopic sinus procedure was done, which confirmed mucormycosis on histopathological examination. After 1 week of conventional amphotericin B and antibiotics, repeat CT brain showed improvement in mucosal thickening and sinusitis. This case is a rare presentation of mucormycosis associated with rapid progression to orbital apex syndrome with brain infarction in a patient with non-ketotic diabetes and COVID-19. Early diagnosis and treatment are essential to prevent further end-organ damage. It is also interesting that there was no angioinvasion and transient periarterial inflammation was attributed to brain infarction.