A5050579303- Optical Imaging and Spectroscopy Techniques
- Brain Tumor Detection and Classification
- Spectroscopy Techniques in Biomedical and Chemical Research
- Remote-Sensing Image Classification
- Photoacoustic and Ultrasonic Imaging
- Advanced Image Fusion Techniques
- Infrared Thermography in Medicine
- CCD and CMOS Imaging Sensors
- Advanced Adaptive Filtering Techniques
- Medical Image Segmentation Techniques
- Computational Physics and Python Applications
- Digital Filter Design and Implementation
- Astronomical Observations and Instrumentation
- Acoustic Wave Phenomena Research
- Blind Source Separation Techniques
- Advanced Memory and Neural Computing
- Numerical methods in inverse problems
- Spectroscopy and Chemometric Analyses
- Retinal Imaging and Analysis
- Advanced Image and Video Retrieval Techniques
- Machine Learning and Data Classification
- Face and Expression Recognition
- Image Enhancement Techniques
- Ferroelectric and Negative Capacitance Devices
- Structural Health Monitoring Techniques
Universidad Politécnica de Madrid
1999-2025
Software (Spain)
2020-2024
Research Institute Hospital 12 de Octubre
2020
Hyperspectral imaging techniques (HSI) do not require contact with patients and are non-ionizing as well non-invasive. As a consequence, they have been extensively applied in the medical field. HSI is being combined machine learning (ML) processes to obtain models assist diagnosis. In particular, combination of these has proven be reliable aid differentiation healthy tumor tissue during brain surgery. ML algorithms such support vector (SVM), random forest (RF) convolutional neural networks...
Hyperspectral imaging sensors have rapidly advanced, aiding in tumor diagnostics for
Hyperspectral (HS) imaging (HSI) techniques have demonstrated to be useful in the medical field characterize tissues without any contact and ionizing patient. Besides, HSI combined with supervised machine learning (ML) algorithms proven an effective technique assist neurosurgeons resect brain tumors. This research looks at effects of hyperparameter optimization on two common ML used for tumor classification: support vector machines (SVM) random forest (RF). Correctly classifying HS data...
Hyperspectral Imaging (HSI) can be used as a non invasive medical diagnostic method when in combination with Machine Learning (ML) algorithms. The significant captured data HSI useful for classifying different types of brain tissues, since they gather reflectance values from band widths below and beyond the visual spectrum. This allows ML algorithms like Support Vector Machines (SVM) Random Forest (RF) to classify tissues such tumors. Predicted results create visualizations support...
<title>Abstract</title> Gliomas constitute a significant challenge in neurosurgery due to their high incidence and poor prognosis. Despite advancements tumor detection techniques using machine learning approach with hyperspectral imaging, accurately distinguishing between healthy tumoral tissues remains challenging. Following this trend, paper introduces Libra, low spectral resolution classifier designed for brain detection, leveraging ensemble enhance classification performance. Evaluated...
Abstract Hyperspectral imaging (HSI) and machine learning (ML) have been employed in the medical field for classifying highly infiltrative brain tumors. Although existing HSI databases of in-vivo human brains are available, they present two main deficiencies. Firstly, amount labeled data is scarce secondly, 3D-tissue information unavailable. To address both issues, we SLIM Brain database, a multimodal image database which provides HS tissue within 400-1000 nm spectrum, as well RGB, depth...
Uyperspectral imaging (HSI) has been adopted during the last years in different applications where material classification plays an important role. This favoured improvement and development of new HS sensors, leading to snapshot cameras, sensor is able acquire images a realtime video fashion, reducing spatial spectral resolution for these cameras. However, although cameras are RAW at high frames per second (FPS) rates, it necessary pre-process them obtain actual image then map. work...
Hyperspectral imaging (HSI) has a crucial role in material classification tasks. The primary type of sensors this field are linescan sensors, offering high spatial and spectral resolution improving the quality captured hypercubes. However, these cameras have inherent problems such as need dedicated hardware to move camera over scene necessity pre-processing raw captures order obtain actual hypercube. This work makes an analysis adaptation stitching techniques apply them medical area. Also,...
Image segmentation tasks often require fully annotated datasets where the boundaries of elements to be identified appear accurately marked. However, such detailed ground truth is hard obtain mainly because it usually involves a time consuming procedure. In biomedical applications, this may imply that medical specialist in charge labelling process can only mark few sparse samples belonging principal interest. context in-vivo brain tumour detection through machine learning techniques and...
In the field of hyperspectral imaging, accurate and reliable data analysis is essential for many applications, including medicine, remote sensing, material science. White calibration a critical step in this process, as it accounts deviations light source intensity spectral distribution. Nevertheless, important to note that specific geometry scene plays crucial role white calibration, affecting angle incidence and, subsequently, measured response. By taking surface normals depth information...
Abstract Significance Hyperspectral imaging sensors have rapidly advanced, aiding in tumor diagnostics for in-vivo brain tumors. Linescan cameras effectively distinguish between pathological and healthy tissue, while snapshot offer a potential alternative to reduce acquisition time. Aim Our research compares linescan hyperspectral tissues chromophores identification. Approach We compared lines-can pushbroom camera using images from 10 patients with various pathologies. Objective comparisons...
With the growth in complexity of image processing algorithms, computational power systems must keep up with requirements. To respond to ubiquitous, spatial or energy constrained environments, developers turn specific embedded systems. In this work, a multiprocessor system-on-chip (MPSoC) ZCU102 Evaluation Kit from Xilinx Zynq Ultrascale+ family is used for hyperspectral (HS) dedicated real-time classification brain tissue. Another main objective work analyze and compare performance realtime...
Hyperspectral imaging analyzed by machine learning algorithms is a powerful tool to classify materials, tissues, molecules and pathogens. By analyzing the electromagnetic spectrum of liquid serum samples, it has been demonstrated that possible predict which patients with head trauma injury will have result on computer tomography. This process being carried out very complex, slow expensive spectrometric techniques. To tackle this problem, study presents simple hyperspectral system allows...
The complexity of the Hyperspectral (HS) imaging-based applications demands faster and more efficient acquisition processing systems. Moreover, HSI technology is being used within medical imaging field, increasing making restrictive requirements implementations. In this work, authors present an methodology for stitching processes when capturing line-scan based HS images. order to verify proposed methodology, a full hardware software setup has been implemented. method tested scanning...
A digital signal processing system with VXI interface has been developed in a C-size board conforming to 1.4 specification, and is made up of two blocks. The first one implements message the second itself, based 49-MHz TMS320C31. includes complete software customize for specific application. gifted real time operating parser building SCPI (standard command programmable instruments) translator. built several modules written C language C30 assembler order improve velocity. utility this realize...
In this abstract a low-cost sound level meter is developed with the help of digital signal processing techniques and an integrated programming environment. The instrument based on use general purpose data acquisition PC card to acquire powerful algorithm filters compute one-third octave band analysis conforming ANSI S1.11-1986 in real time. objective made possible thanks optimized routines from Intel Signal Processing Library. library instrumentation specific environment such as...