Fractional derivative based weighted skip connections for satellite image road segmentation.

Segmentation of a road portion from a satellite image is challenging due to its complex background, occlusion, shadows, clouds, and other optical artifacts. One must combine both local and global cues for an accurate and continuous/connected road network extraction. This paper proposes a model using fractional derivative-based weighted skip connections on a densely connected convolutional neural network for road segmentation. Weights corresponding to the skip connections are determined using Grunwald-Letnikov fractional derivative. Fractional derivatives being non-local in nature incorporates memory into the system and thereby combine both local and global features. Experiments have been performed on two open source widely used benchmark databases viz. Massachusetts Road database (MRD) and Ottawa Road database (ORD). Both these datasets represent different road topography and network structure including varying road widths and complexities. Result reveals that the proposed system demonstrated better performance than the other state-of-the-art methods by achieving an F1-score of 0.748 and the mIoU of 0.787 at fractional order 0.4 on the MRD and a mIoU of 0.9062 at fractional order 0.5 on the ORD.

DroneAttention: Sparse weighted temporal attention for drone-camera based activity recognition.

Human activity recognition (HAR) using drone-mounted cameras has attracted considerable interest from the computer vision research community in recent years. A robust and efficient HAR system has a pivotal role in fields like video surveillance, crowd behavior analysis, sports analysis, and human-computer interaction. What makes it challenging are the complex poses, understanding different viewpoints, and the environmental scenarios where the action is taking place. To address such complexities, in this paper, we propose a novel Sparse Weighted Temporal Attention (SWTA) module to utilize sparsely sampled video frames for obtaining global weighted temporal attention. The proposed SWTA is comprised of two parts. First, temporal segment network that sparsely samples a given set of frames. Second, weighted temporal attention, which incorporates a fusion of attention maps derived from optical flow, with raw RGB images. This is followed by a basenet network, which comprises a convolutional neural network (CNN) module along with fully connected layers that provide us with activity recognition. The SWTA network can be used as a plug-in module to the existing deep CNN architectures, for optimizing them to learn temporal information by eliminating the need for a separate temporal stream. It has been evaluated on three publicly available benchmark datasets, namely Okutama, MOD20, and Drone-Action. The proposed model has received an accuracy of 72.76%, 92.56%, and 78.86% on the respective datasets thereby surpassing the previous state-of-the-art performances by a margin of 25.26%, 18.56%, and 2.94%, respectively.

Air Quality Prediction Using the Fractional Gradient-Based Recurrent Neural Network.

In this study, the air quality index (AQI) of Indian cities of different tiers is predicted by using the vanilla recurrent neural network (RNN). AQI is used to measure the air quality of any region which is calculated on the basis of the concentration of ground-level ozone, particle pollution, carbon monoxide, and sulphur dioxide in air. Thus, the present air quality of an area is dependent on current weather conditions, vehicle traffic in that area, or anything that increases air pollution. Also, the current air quality is dependent on the climate conditions and industrialization in that area. Thus, the AQI is history-dependent. To capture this dependency, the memory property of fractional derivatives is exploited in this algorithm and the fractional gradient descent algorithm involving Caputo's derivative has been used in the backpropagation algorithm for training of the RNN. Due to the availability of a large amount of data and high computation support, deep neural networks are capable of giving state-of-the-art results in the time series prediction. But, in this study, the basic vanilla RNN has been chosen to check the effectiveness of fractional derivatives. The AQI and gases affecting AQI prediction results for different cities show that the proposed algorithm leads to higher accuracy. It has been observed that the results of the vanilla RNN with fractional derivatives are comparable to long short-term memory (LSTM).

A Study on Clinical Profile of Hemifacial Spasm in India and the Therapeutic Response to Botulinum Toxin Type A Injection as well as Pre and Postinjection Quality of Life.

BACKGROUND: Hemifacial spasm (HFS) is a distressing, involuntary, irregular tonic-clonic contraction of the facial muscles innervated by the seventh cranial nerve. It affects the quality of life. Botulinum toxin is a preferred symptomatic treatment option for the condition. However, there is a lack of study in the Indian scenario. Therefore, we observed the demographic profile, clinical spectrum, therapeutic response, and adverse effects of botulinum toxin and assessed the quality of life in the pre and postinjection phases in our subjects with HFS. MATERIALS AND METHODS: The study design is a prospective open-label observational study. Consecutive cases of HFS were selected from the general neurology outpatient department (OPD) and movement disorder clinic of a medical college hospital in Eastern India. Clinical and relevant neuroimaging studies excluded mimickers and secondary causes of HFS. Institutional Ethics Committee's permission was obtained. Informed consent was taken from patients before botulinum toxin injection. The pre and postinjection assessment tools were spasm rate for a specific period of time, quantification of facial asymmetry, widening palpebral fissure by visual analog scale, Jankovic disability rating scale, HFS-7 scale, and videography. RESULTS: A total of 250 cases of HFS (F:M = 138:112) were studied. The mean age of presentation was 47 years. The mean dose of botulinum toxin injection was 24.2 units per patient. The mean duration of improvement was 4 months. The spasm frequency was decreased by 90%, and the facial asymmetry was improved by 86%. The improvement in quality of life was 86%. Local adverse effects are seen in 10.4% of cases, and all were reversible. CONCLUSION: This is one of the largest studies on the effects of botulinum toxin in subjects with HFS in the Indian population. Periodic injection of botulinum toxin is a safe and effective therapy for subjects with HFS. There is a significant improvement in the quality of life following botulinum toxin therapy in subjects with HFS. Adverse effects were local, mild, well-tolerated, and reversible.

CSITime: Privacy-preserving human activity recognition using WiFi channel state information.

Human activity recognition (HAR) is an important task in many applications such as smart homes, sports analysis, healthcare services, etc. Popular modalities for human activity recognition involving computer vision and inertial sensors are in the literature for solving HAR, however, they face serious limitations with respect to different illumination, background, clutter, obtrusiveness, and other factors. In recent years, WiFi channel state information (CSI) based activity recognition is gaining momentum due to its many advantages including easy deployability, and cost-effectiveness. This work proposes CSITime, a modified InceptionTime network architecture, a generic architecture for CSI-based human activity recognition. We perceive CSI activity recognition as a multi-variate time series problem. The methodology of CSITime is threefold. First, we pre-process CSI signals followed by data augmentation using two label-mixing strategies - mixup and cutmix to enhance the neural network's learning. Second, in the basic block of CSITime, features from multiple convolutional kernels are concatenated and passed through a self-attention layer followed by a fully connected layer with Mish activation. CSITime network consists of six such blocks followed by a global average pooling layer and a final fully connected layer for the final classification. Third, in the training of the neural network, instead of adopting general training procedures such as early stopping, we use one-cycle policy and cosine annealing to monitor the learning rate. The proposed model has been tested on publicly available benchmark datasets, i.e., ARIL, StanWiFi, and SignFi datasets. The proposed CSITime has achieved accuracy of 98.20%, 98%, and 95.42% on ARIL, StanWiFi, and SignFi datasets, respectively, for WiFi-based activity recognition. This is an improvement on state-of-the-art accuracies by 3.3%, 0.67%, and 0.82% on ARIL, StanWiFi, and SignFi datasets, respectively. In lab-5 users' scenario of the SignFi dataset, which has the training and testing data from different distributions, our model achieved accuracy was 2.17% higher than state-of-the-art, which shows the comparative robustness of our model.

Analysis of clinical and radiological features of tuberculosis associated pneumothorax.

AIM: To investigate the demographic and clinico-radiological characteristic of patients of tuberculosis presented with pneumothorax in relation to the patients of active pulmonary tuberculosis. MATERIAL AND METHODS: The retrospective study was conducted between January 2013 to June 2014 and records of 78 patients of pulmonary tuberculosis with pneumothorax (TP) and 156 patients of pulmonary tuberculosis without pneumothorax (NPT) were analyzed. Demographic, etiologic, clinical, radiographic, and outcome data were collected. RESULTS: The mean age of tuberculous pneumothorax patients was 38.0±14.3 years and that of non-pneumothorax was 39.5±12.3 years. Most common presenting clinical feature was cough (76.9%) followed by dyspnoea (74.4%), chest pain (64.0%) and fever (56.4%) in TP patients. Chest radiograph showed cavity in 38 (48.7%) TP patients followed by consolidation in 32 (41.0%) patients and infiltration in 33 (42.3%) patients. All the TP patients had undergone underwater seal intercostal tube drainage procedure for the management of pneumothorax. The mean duration between chest drain insertion and removal was 17.14±6.37 days. Twenty-five (32.1%) of the cases developed chest tube drainage related complications. CONCLUSION: Patients of tuberculous pneumothorax required prolonged period of chest tube drainage and usually showed good response to the treatment.

Bronchial thermoplasty: a non-pharmacological approach.

INTRODUCTION: Asthma is a chronic inflammatory disorder of the airway characterized by the episodic symptoms of breathlessness, wheezes and cough. Even with the use of maximum anti-asthmatic pharmacological treatment sometimes it remains uncontrolled. For such patients, bronchial thermoplasty is the new mode of treatment. OBJECTIVE: To review published article on bronchial thermoplasty. METHODS: We identified 102 English articles on PubMed, and 56 were excluded by the abstract. The remaining articles were retrieved for full-text detailed evaluation by authors, and 28 relevant articles were selected for final review. RESULTS: Bronchial thermoplasty is the radiofrequency ablation of the airway smooth muscle with the help of flexible fiberoptic bronchoscope. It reduces the smooth muscle mass of the bronchial wall and decreases its contractility. CONCLUSION: Bronchial thermoplasty causes improvement in the quality of life, and causes reduction in the emergency room visit and exacerbation due to asthma. Long-term safety has been established by various prospective studies.

Improvement in quality of life in pulmonary tuberculosis patients: a prospective study.

A cohort of 50 newly diagnosed patients with pulmonary tuberculosis was prospectively studied and compared with controls to assess and quantify their quality of life using the World Health Organization's (WHO) Quality of Life (QOL)-BREF score before and after a Directly Observed Therapy Short (DOTS) course. It was concluded that health-related quality of life is impaired by tuberculosis and shows significant improvement with the DOTS treatment.

Tuberculous lymphadenopathy: Experience from the referral center of Northern India.

BACKGROUND: Lymphadenopathy is the common form of extrapulmonary tuberculosis (TB) in the developing country like India. The aim of the study was to assess the clinical and the pathological features of tuberculous peripheral lymphadenopathy along with the effect of the antitubercular drugs on its clinical course. MATERIALS AND METHODS: A prospective study was carried out in the department of pulmonary medicine from July 2013 to June 2014. Clinico-demographical and pathological characteristics of the patients of tuberculous lymphadenopathy were studied. RESULTS: Two and four cases, 84 (41.2%) male and 120 (58.8%) females were diagnosed as the cases of tuberculous lymphadenopathy. Mean age of the patients were 28.45 ± 12.83. Palpable mass (100%) was the most common presenting feature. Cervical lymphadenopathy (92.6%) was the most common presentation of peripheral tuberculous lymphadenopathy. Most common cytological pattern seen was epithelioid granuloma with caseous necrosis (32.84%) followed by epithelioid granuloma without caseous necrosis on fine needle aspiration cytology (FNAC). Directly observed treatment short course (DOTS) for TB was effective in treating tuberculous lymphadenopathy. CONCLUSION: Tuberculous lymphadenopathy affects female more often than males. FNAC is the effective mean in diagnosing tuberculous lymphadenopathy. DOTS is the effective mean for treating tuberculous lymphadenopathy.

Etiology of hemoptysis: A retrospective study from a tertiary care hospital from northern Madhya Pradesh, India.

OBJECTIVE: To evaluate the various etiologies of hemoptysis and outcome in an Indian cohort. MATERIAL AND METHODS: Retrospective analysis of patients admitted with complaints of hemoptysis in the department of Pulmonary Medicine between April 2010 and March 2013. We categorized the patients according to various etiologies and according to the grades of hemoptysis. RESULTS: Three hundred and forty-six patients were included in the study. Of these, 214 (67%) were men and 142 (33%) were women. Tuberculosis (79.2%) accounted for majority of cases of hemoptysis. Other causes of hemoptysis were lung cancer (7.2%), bronchitis (4.6%), and bronchiectasis (3.5%). Moderate grade (73.4%) of hemoptysis was the frequent mode of presentation. CONCLUSION: Hemoptysis is the most frequent presentation of tuberculosis in India. It may be the presentation of healed pulmonary tuberculosis. Moderate amount of hemoptysis is commonly seen in the general population.

Clinicoradiological Profile of Lower Lung Field Tuberculosis Cases among Young Adult and Elderly People in a Teaching Hospital of Madhya Pradesh, India.

Aim. To study the clinical and radiological features of lower lung field tuberculosis (LLFTB) in relation to the patients of nonlower lung field tuberculosis (non-LLFTB). Material and Methods. All the patients of lower lung field tuberculosis defined by the lesions below an arbitrary line across the hila in their chest X-rays were included in the study. Their sputum for acid fast bacilli, HIV, blood sugar, and other relevant investigations were performed. Results. The total of 2136 cases of pulmonary tuberculosis was studied. Among them 215 (10%) cases of patients were diagnosed as the case of lower lung field tuberculosis. Females (62%) were more commonly affected. Most common clinical feature in non-LLFTB was cough (69%) followed by fever (65%), chest pain (54.7%), and weight loss (54.4%). Chest X-ray showed predominance of right side (60.9%) in cases of LLFTB. The relative risk of having the LLFTB in diabetes patients, HIV seropositive patients, end stage renal disease patients, and patients on corticosteroid therapy was high. Conclusion. Lower lung field tuberculosis is not an uncommon entity. It is more common in diabetes, HIV positive, end stage renal disease, and corticosteroid treated patients. Clinical and radiological features are different from upper lobe tuberculosis patients.

Effect of land use land cover change on soil erosion potential in an agricultural watershed.

Universal soil loss equation (USLE) was used in conjunction with a geographic information system to determine the influence of land use and land cover change (LUCC) on soil erosion potential of a reservoir catchment during the period 1989 to 2004. Results showed that the mean soil erosion potential of the watershed was increased slightly from 12.11 t ha(-1) year(-1) in the year 1989 to 13.21 t ha(-1) year(-1) in the year 2004. Spatial analysis revealed that the disappearance of forest patches from relatively flat areas, increased in wasteland in steep slope, and intensification of cultivation practice in relatively more erosion-prone soil were the main factors contributing toward the increased soil erosion potential of the watershed during the study period. Results indicated that transition of other land use land cover (LUC) categories to cropland was the most detrimental to watershed in terms of soil loss while forest acted as the most effective barrier to soil loss. A p value of 0.5503 obtained for two-tailed paired t test between the mean erosion potential of microwatersheds in 1989 and 2004 also indicated towards a moderate change in soil erosion potential of the watershed over the studied period. This study revealed that the spatial location of LUC parcels with respect to terrain and associated soil properties should be an important consideration in soil erosion assessment process.