A5082499287- Retinal Diseases and Treatments
- Retinal Imaging and Analysis
- Retinal and Optic Conditions
- Memory and Neural Mechanisms
- Primate Behavior and Ecology
- Face Recognition and Perception
- Zebrafish Biomedical Research Applications
- Visual perception and processing mechanisms
- Retinal and Macular Surgery
- Glaucoma and retinal disorders
- Digital Imaging for Blood Diseases
- Advanced Graph Neural Networks
- Plant biochemistry and biosynthesis
- Imbalanced Data Classification Techniques
- Gaze Tracking and Assistive Technology
- Corneal surgery and disorders
- Plant Reproductive Biology
- Philosophy and History of Science
- Neural and Behavioral Psychology Studies
- Plant and animal studies
- Topic Modeling
- Ocular Oncology and Treatments
- Tactile and Sensory Interactions
- Aortic Thrombus and Embolism
- Data Visualization and Analytics
Indian Institute of Science Bangalore
2020-2025
Little Flower Hospital & Research Centre
2014-2023
Lakeshore Hospital
2014
University of Kerala
1981
Diabetic retinopathy (DR) is one of the most common retinal diseases that cause preventable blindness in diabetic patients. The timely screening and grading images minimize possibility vision loss. However, manual images, for detecting micro lesions early stages DR, time-consuming. This paper proposes an ensemble deep convolutional neural network (CNN) models accurate detection DR using fundus images. Each input image divided into four patches at first stage passed on to pre-trained CNN...
Diabetic retinopathy (DR) at risk of vision loss (referable DR) needs to be identified by retinal screening and referred an ophthalmologist. Existing automated algorithms have mostly been developed from images acquired with high cost mydriatic cameras cannot applied in the settings used most low- middle-income countries. In this prospective multicentre study, we a deep learning system (DLS) that detects referable DR using handheld non-mydriatic fundus camera non-technical field workers 20...
National estimates of the prevalence vision impairment and blindness in people with diabetes are required to inform resource allocation. People more susceptible conditions such as diabetic retinopathy that can impair vision; however, these often missed national studies. This study aims determine risk factors India. Data from SMART-India study, a cross-sectional survey coverage 42 147 Indian adults aged 40 years older, collected using complex sampling design, were used obtain nationally...
ABSTRACT Monkeys are widely used as model organisms for human vision and cognition. While their anatomy physiology strongly correspond to humans, it is unclear what extent perception matches with ours. Previous studies have evaluated specific aspects of perception, after extensive training on customized tasks that could altered perception. To resolve these issues, we trained monkeys perform an oddball visual search task natural images, tested them carefully controlled images detect a variety...
Screening effectively identifies patients at risk of sight-threatening diabetic retinopathy (STDR) when retinal images are captured through dilated pupils. Pharmacological mydriasis is not logistically feasible in non-clinical, community DR screening, where acquiring gradable using handheld devices exhibits high technical failure rates, reducing STDR detection. Deep learning (DL) based gradability predictions acquisition could prompt device operators to recapture insufficient quality images,...
Macaque monkeys are widely used to study vision. In the traditional approach, brought into a lab perform visual tasks while they restrained obtain stable eye tracking and neural recordings. Here, we describe novel environment cognition in more natural setting as well other social behaviors. We designed naturalistic with an integrated touchscreen workstation that enables high-quality unrestrained monkeys. this train on challenging same-different task. also show can reveal interesting As proof...
Diabetic retinopathy is one of the most common retinal diseases in diabetic patients, resulting avoidable blindness. Therefore, early detection and grading images are crucial to reduce risk vision loss. Also, DR an automated manner critical for effective care. The objective this study classify evaluate fundus into two levels. In current scenario, compared feature-based image classification techniques, convolutional neural networks have a higher efficiency. This work presents comparative...
Language models, being at the heart of many NLP problems, are always great interest to researchers. Neural language models come with advantage distributed representations and long range contexts. With its particular dynamics that allow cycling information within network, `Recurrent neural network' (RNN) becomes an ideal paradigm for modeling. Long Short-Term Memory (LSTM) architecture solves inadequacies standard RNN in modeling long-range In spite a plethora variants, possibility add...
Diabetic retinopathy is an irreversible eye disease that a prominent cause of blindness globally. For proper diagnosis and vision protection, it essential to detect DR early accurately. In the current scenario challenging in automated manner using fundus images. This work investigates performance deep learning based CNN models effective classification. We explore three popular (Xception, InceptionV3, ResNet50) for categorization this retinal disease. A comparative study pre-trained are...
Abstract A case of malignant pleomorphic adenoma the left parotid gland which metastasized to kidney and multiple subcutaneous sites is presented. The patient presented as an acute emergency due spontaneous rupture containing metastasis. All distant metastatic lesions were confined right side body.
ABSTRACT Macaque monkeys are widely used to study the neural basis of cognition. In traditional approach, monkey is brought into a lab perform tasks while it restrained obtain stable gaze tracking and recordings. This unnatural setting prevents studying brain activity during natural, social complex behaviors. Here, we designed naturalistic environment with an integrated behavioral workstation that enables task training viable in freely moving monkeys. We this facility train on challenging...
To report the case of a 52-year-old female with recent complaint decreased vision associated well-defined depigmented lesion in macula right eye. She had history surgery for pituitary microadenoma and breast lump excision. Based on spectral-domain optical coherence tomography, fundus fluorescein angiography, indocyanine green B-scan ultrasonography, diagnosis eye choroidal osteoma was made. Subretinal fluid (SRF) noted absence obvious classic neovascularization (CNV). The patient given...
Cytological investigations on Camellia sinensis, C. assamica, and their hybrid (Jat) were undertaken in the present study.The somatic chromosome number (2n=30) observed species sinensis assamica is conformity with earlier reports. Also meiosis pollen mother cells was regular these species. However irregular China-Assam hybrid. Localized unpairing at pachytene, stickiness resulting apparent higher association, laggards, sticky bridges etc. variations of daughter nuclei telophase II...
Cholesterol embolism is a rare complication of percutaneous vascular procedures. It usually diagnosed from the typical cutaneous manifestations and delayed development renal failure. Here we report patient who developed incessant vomiting following procedure later cholesterol embolism. Even though gastrointestinal may be seen up to 20% patients, this first case in which symptoms are as initial presentation
When a spiky object is occluded, we have no trouble believing that the hidden portions are also spiky. Whether this process occurs automatically in perception, or requires cognitive reasoning unclear. Here, devised novel “free-choice” paradigm to reveal what monkeys see behind an occluded display. On 80% of trials, denoted as unambiguous two items (sample and test) were shown with intervening delay, trained make “same” “different” response. These trials consisted unoccluded displays...
When a spiky object is occluded, we have the compelling percept that its features continue behind occluder. Although many real-world objects contain complex features, it unclear how their are amodally completed and whether this process automatic. To end, asked participants to search for oddball targets among distractors similarity relations in visual between occluded displays match better with global or local completions of these displays. In Experiment 1, when curved/straight corners were...
We can easily recognize objects even when they are occluded. Does our perception complete the occluded edges linearly or it also employ more sophisticated extrapolation? To investigate this, we created displays where each shape had two possible completions. These identical contours in unoccluded portion but differed portion, such that one appeared to be likely behind occluder, and other was unlikely. Across sets of displays, manipulated either curvature edge (‘curvature’ set) numerosity...