A5059110757- Metabolomics and Mass Spectrometry Studies
- COVID-19 diagnosis using AI
- Spectroscopy Techniques in Biomedical and Chemical Research
- Radiomics and Machine Learning in Medical Imaging
- Mass Spectrometry Techniques and Applications
- Advanced Chemical Sensor Technologies
- EEG and Brain-Computer Interfaces
- Spectroscopy and Chemometric Analyses
- Autism Spectrum Disorder Research
- Emotion and Mood Recognition
- AI in cancer detection
- Molecular Biology Techniques and Applications
- Photoacoustic and Ultrasonic Imaging
- Advanced Image Fusion Techniques
- Infant Health and Development
- Social Robot Interaction and HRI
- Child Development and Digital Technology
- Radiology practices and education
- Lung Cancer Diagnosis and Treatment
Thermo Fisher Scientific (United States)
2025
University of Tennessee at Knoxville
2021-2024
Texas State University
2018-2019
Abstract This study aims to assess the impact of domain shift on chest X-ray classification accuracy and analyze influence ground truth label quality demographic factors such as age group, sex, year. We used a DenseNet121 model pre-trained MIMIC-CXR dataset for deep learning-based multi-label using labels from radiology reports extracted CheXpert CheXbert Labeler. compared performance 14 Veterans Healthcare Administration (VA-CXR). The validation assessment across various NLP extraction...
PurposeDeep learning (DL) models have received much attention lately for their ability to achieve expert-level performance on the accurate automated analysis of chest X-rays (CXRs). Recently available public CXR datasets include high resolution images, but state-of-the-art are trained reduced size images due limitations graphics processing unit memory and training time. As computing hardware continues advance, it has become feasible train deep convolutional neural networks high-resolution...
In this paper, an initial work of a research is discussed which to teach young autistic children recognizing human facial expression with the help computer vision and image processing. This paper mostly discusses recognition using deep convolutional neural network. The Kaggle's FER2013 dataset has been used train experiment network model. Once satisfactory result achieved, modified pictures four different lighting conditions each these datasets again trained same necessary for end goal...
Recent advancements in molecular Mass Spectrometry Imaging (MSI) have sparked interest integrating high spatial resolution methods with mass-spectrometry-based chemical imaging. Fusion-based algorithms proven effective generating spatial-resolution mass spectra. However, a significant challenge stems from the differing physical mechanisms underlying image generation and data upsampling techniques, potentially leading to discrepancies integrated information channels. Integrating constraints...
Prostate cancer is one of the most common cancers globally and second in male population US. Here we develop a study based on correlating hematoxylin eosin (H&E)-stained biopsy data with MALDI mass-spectrometric imaging corresponding tissue to determine cancerous regions their unique chemical signatures variations predicted original pathological annotations. We obtain features from high-resolution optical micrographs whole slide H&E stained through deep learning spatially register them mass...
In this paper, continued work of a research project is discussed whose end goal to build mobile device application that can teach children with ASD (Autism Spectrum Disorder) recognize human facial expressions utilizing computer vision and image processing. This paper discusses the intermediate expression recognition approach using deep convolutional neural network (DCNN) images from different angles. The Karolinska Directed Emotional Faces (KDEF) dataset has been used train test DCNN model....
Deep learning (DL) has become an indispensable tool in hyperspectral data analysis, automatically extracting valuable features from complex, high-dimensional datasets. Super-resolution reconstruction, essential aspect of data, involves enhancing spatial resolution, particularly relevant to low-resolution data. Yet, the pursuit super-resolution analysis is fraught with challenges, including acquiring ground truth high-resolution for training, generalization, and scalability. The pressing...
In this paper, continued work of a research project is discussed which achieved the end goal - to build mobile device application that can teach children with Autism Spectrum Disorder (ASD) recognize human facial expressions utilizing computer vision and image processing. Universally, there are seven categories: angry, disgust, happy, sad, fear, surprise, neutral. To all these predict current mood person difficult task for child. A child ASD, problem presents itself in more sophisticated...
Abstract Deep learning models have received much attention lately for their ability to achieve expert-level performance on the accurate automated analysis of chest X-rays. Although publicly available X-ray datasets include high resolution images, most are trained reduced size images due limitations GPU memory and training time. As compute capability continues advance, it will become feasible train large convolutional neural networks high-resolution images. To verify that this lead increased...
Objectives: This study aims to assess the impact of domain shift on chest X-ray classification accuracy and analyze influence ground truth label quality demographic factors such as age group, sex, year. Materials Methods: We used a DenseNet121 model pretrained MIMIC-CXR dataset for deep learning-based multilabel using labels from radiology reports extracted CheXpert CheXbert Labeler. compared performance 14 Veterans Healthcare Administration (VA-CXR). The VA-CXR comprises over 259k images...
Abstract Prostate cancer is one of the most common cancers globally and second in male population US. Here we develop a study based on correlating H&E-stained biopsy data with MALDI mass-spectrometric imaging corresponding tissue to determine cancerous regions their unique chemical signatures, variation predicted original pathological annotations. We spatially register features obtained through deep learning from high-resolution optical micrographs whole slide H&E stained MSI...
An abstract is not available for this content so a preview has been provided. As you have access to content, full PDF via the ‘Save PDF’ action button.