Sharmila Nageswaran

Bio: Sharmila Nageswaran is an academic researcher from VIT University. The author has contributed to research in topics: Foot (unit) & Image processing. The author has an hindex of 5, co-authored 21 publications receiving 43 citations.

Lung Cancer Classification and Prediction Using Machine Learning and Image Processing

Abstract: Lung cancer is a potentially lethal illness. Cancer detection continues to be a challenge for medical professionals. The true cause of cancer and its complete treatment have still not been discovered. Cancer that is caught early enough can be treated. Image processing methods such as noise reduction, feature extraction, identification of damaged regions, and maybe a comparison with data on the medical history of lung cancer are used to locate portions of the lung that have been impacted by cancer. This research shows an accurate classification and prediction of lung cancer using technology that is enabled by machine learning and image processing. To begin, photos need to be gathered. In the experimental investigation, 83 CT scans from 70 distinct patients were utilized as the dataset. The geometric mean filter is used during picture preprocessing. As a consequence, image quality is enhanced. The K-means technique is then used to segment the images. The part of the image may be found using this segmentation. Then, classification methods using machine learning are used. For the classification, ANN, KNN, and RF are some of the machine learning techniques that were used. It is found that the ANN model is producing more accurate results for predicting lung cancer.

Design of mechanical interface to re-distribute excess pressure to prevent the formation of decubitus ulcers in bed ridden patients

Abstract: Pressure ulcers are the major problem in the stroke management and rehabilitation. Prevention of pressure ulcer is of keen interest and is achieved by frequently changing the position of patient on the mattress. However, the care needs to be intensive to address this issue; else it would lead to pressure ulcer or bed sores formation. Skin surface over the bony prominences provide comparatively more pressure than the other regions. Therefore they are called as pressure vulnerable regions. Skin over these regions is more susceptible for formation of ulcers. An engineering approach is needed to shift the accumulating pressure from the pressure vulnerable regions. Although pressure sensed in these region would be more than that of which sensed in other regions, shifting protocol has to be designed to channelize or to grade the pressure shift in order to avoid any injuries to the non pressure vulnerable region. This paper aims at devising one such protocol using MATLab and thereby designing the layout of mattress using Pro/Engineer: the number of partitions needed to cover the entire surface of the skin that is in contact with the mattress.

Analysis of pressure relieving mechanisms for the prevention of decubitus ulcers

Abstract: Pressure Ulcers, medically coined as 'decubitus ulcers' have been a major threat for rehabilitation of bed ridden patients for about a century, as these ulcers eventually leads to fatality Managing the ulcer is very challenging so the best option being their prevention There were two broad theories behind the etiology of its formation namely neurogenic and mechanical Therefore intervention should also be neurogenic and mechanical Neurogenic intervention is to prevent the chance of occurrence of stroke and other neurological pathologies while mechanical intervention is to prevent the accumulation of excess pressure under ulcer vulnerable regions The skin over bony prominences are said to be pressure vulnerable regions There were various technology developed to prevent the formation of ulcers This paper will review existing technology and will add more on what is needed for early prevention

A novel way to acquire foot contour measurements of remotely located patients having foot deformities

Abstract: In developing countries, patients with disabilities are often deprived of getting treated at rehabilitation centres with modern equipments and technologies. In India as per the 2011 census about 2.21% of the population suffer with disabilities due to either visual, auditory, speech or locomotor dysfunctions. Locomotor disability is seen predominantly in about 20.27% of patients amongst the various other disabilities. Measurement of the foot deformities is very essential for the fabrication of orthosis. The techniques of foot measurement that are presently used needs either the patient's foot outline measurement to be sent to the rehabilitation centre or cast measurement to be taken using plaster of paris mould. The need of the hour is for a technique that is quick and doesn't require the patient to visit the rehabilitation centre. This paper suggests a method in which the digital photograph of the foot outline of a remotely located patient would help to acquire the required dimensions of the foot. The developed technique of measuring foot outline has been validated by comparing n=103 pairs of digital photographs of patient's foot outline with that of the manual method of measurement. Error in measurement using this technique was found to be 2.83 ±0.39mm. Also Pearson correlation coefficient of measured foot length and width were 0.991±0.005 for a 95% Confidence Interval and was very significant (p<0.001). Acquiring foot measurement of remotely located patients through these methods can help the clinical experts in the rehabilitation centre fabricate appropriate foot wear or orthotic device for the patient.

Prevention and treatment of pressure ulcers

Abstract: Pressure is the primary pathogenic factor in the development of ulcers but other major factors are shearing forces, friction and moisture. Significant risk factors are immobility, nutritional status and age-related diseases. Complications of pressure ulcers can be life threatening. The authors wish to stress preventive measures and also recommend a therapeutic approach based on the grade of ulcer present. Prevention is best achieved by identification of high risk patients, alleviation of causative and predisposing factors and early detection of ischemic skin changes. The treatment includes local wound care that eliminates necrotic tissue, decreases bacterial load and provides a physiologic environment allowing the wound to heal, and systemic treatment: adequate nutrition, correction of underlying illnesses and systemic antibiotics (in case of sepsis, cellulitis, osteomyelitis or the prevention of bacterial endocarditis).

Pressure ulcers: the great insult.

Abstract: Pressure ulcers are a common and frustrating problem. Pressure ulcers increase demands on health care resources and are sometimes a source of malpractice litigation. Skin breakdown, often an iatrogenic complication of hospitalization, increases the length of stay and contributes to mortality and morbidity. Long-term care facilities are under increasing regulatory pressure to reduce rates of pressure ulcer occurrence. The process-outcome link continues to escape us. Processes of care seem disjointed. Numerous studies show a failure to implement what we know. When pressure ulcer risk is identified, preventive measures often are not implemented. The literature is replete with reports of quality improvement activities that enumerate multiple opportunities to improve care related to pressure ulcers. Various quality improvement strategies for pressure ulcer prevention and management have been produced, but recommendations are not always applied to practice. When studies compared various outcomes before and after implementation of guidelines, most of the evidence was clinical audit data. Overall, active strategies were associated with better outcomes and passive strategies with poorer ones. Baier et al reported improvement in processes of care after using a structured quality improvement approach in the long-term care setting. Targeted education sessions were common to all studies reporting successful outcomes. Multidisciplinary wound care teams that conduct rounds at the bedside are highly recommended to enhance patient outcomes.Functioning interdisciplinary teams clearly represent an important approach to error reduction. To close the gap between risk identification and pressure ulcer prevention, we should develop active multidisciplinary wound care teams and \"Strive for Six Sigma in Pressure Ulcer Care\".

Competitive Learning in a Spiking Neural Network: Towards an Intelligent Pattern Classifier.

Abstract: One of the modern trends in the design of human-machine interfaces (HMI) is to involve the so called spiking neuron networks (SNNs) in signal processing. The SNNs can be trained by simple and efficient biologically inspired algorithms. In particular, we have shown that sensory neurons in the input layer of SNNs can simultaneously encode the input signal based both on the spiking frequency rate and on varying the latency in generating spikes. In the case of such mixed temporal-rate coding, the SNN should implement learning working properly for both types of coding. Based on this, we investigate how a single neuron can be trained with pure rate and temporal patterns, and then build a universal SNN that is trained using mixed coding. In particular, we study Hebbian and competitive learning in SNN in the context of temporal and rate coding problems. We show that the use of Hebbian learning through pair-based and triplet-based spike timing-dependent plasticity (STDP) rule is accomplishable for temporal coding, but not for rate coding. Synaptic competition inducing depression of poorly used synapses is required to ensure a neural selectivity in the rate coding. This kind of competition can be implemented by the so-called forgetting function that is dependent on neuron activity. We show that coherent use of the triplet-based STDP and synaptic competition with the forgetting function is sufficient for the rate coding. Next, we propose a SNN capable of classifying electromyographical (EMG) patterns using an unsupervised learning procedure. The neuron competition achieved via lateral inhibition ensures the "winner takes all" principle among classifier neurons. The SNN also provides gradual output response dependent on muscular contraction strength. Furthermore, we modify the SNN to implement a supervised learning method based on stimulation of the target classifier neuron synchronously with the network input. In a problem of discrimination of three EMG patterns, the SNN with supervised learning shows median accuracy 99.5% that is close to the result demonstrated by multi-layer perceptron learned by back propagation of an error algorithm.

Experimental and numerical diagnosis of fatigue foot using convolutional neural network.

Abstract: Fatigue is an essential criterion for physiotherapy in injured athletes. Muscle fatigue mechanism also is a crucial matter in designing a workout program. It is mainly related to physical injury, c...