A5076935643- Retinal Imaging and Analysis
- Glaucoma and retinal disorders
- Advanced Neural Network Applications
- Retinal Diseases and Treatments
- Advanced X-ray and CT Imaging
- Retinal and Optic Conditions
- 3D Printing in Biomedical Research
- Medical Imaging Techniques and Applications
- Cell Image Analysis Techniques
- Digital Imaging for Blood Diseases
- Viral Infectious Diseases and Gene Expression in Insects
- Adversarial Robustness in Machine Learning
- Anomaly Detection Techniques and Applications
- Innovative Microfluidic and Catalytic Techniques Innovation
- Microfluidic and Bio-sensing Technologies
- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging and Analysis
- Pluripotent Stem Cells Research
- Image Processing Techniques and Applications
- Corneal surgery and disorders
- Radiation Dose and Imaging
- Image and Object Detection Techniques
- Optical Coherence Tomography Applications
- Nuclear Physics and Applications
- Advanced Fluorescence Microscopy Techniques
University College London
2010-2017
John Wiley & Sons (United States)
2014
Bioengineering Center
2014
National Center for Genetic Engineering and Biotechnology
2014
École Polytechnique Fédérale de Lausanne
2006-2008
The quantitative determination of key adherent cell culture characteristics such as confluency, morphology, and density is necessary for the evaluation experimental outcomes to provide a suitable basis establishment robust protocols. Automated processing images acquired using phase contrast microscopy (PCM), an imaging modality widely used visual inspection cultures, could enable non-invasive these characteristics. We present image-processing approach that accurately detects cellular objects...
The early detection of glaucoma is essential in preventing visual impairment. Artificial intelligence (AI) can be used to analyze color fundus photographs (CFPs) a cost-effective manner, making screening more accessible. While AI models for from CFPs have shown promising results laboratory settings, their performance decreases significantly real-world scenarios due the presence out-of-distribution and low-quality images. To address this issue, we propose Intelligence Robust Glaucoma...
Non-intrusive inspection systems based on X-ray radiography techniques are routinely used at transport hubs to ensure the conformity of cargo content with supplied shipping manifest. As trade volumes increase and regulations become more stringent, manual by trained operators is less viable due low throughput. Machine vision can assist in their task automating parts workflow. Since cars involved trafficking, export fraud, tax evasion schemes, they represent an attractive target for automated...
We review the relatively immature field of automated image analysis for X-ray cargo imagery. There is increasing demand methods that can assist in inspection and selection containers, due to ever-growing volumes traded concerns cust oms- security-related threats are being smuggled across borders by organised crime terrorist networks. split into classical pipeline preprocessing understanding. Preprocessing includes: manipulation; quality improvement; Threat Image Projection (TIP); material...
Abstract Oxygen plays a key role in stem cell biology as signaling molecule and an indicator of energy metabolism. Quantification cellular oxygen kinetics, i.e. the determination specific uptake rates (sOURs), is routinely used to understand metabolic shifts. However current methods determine sOUR adherent cultures rely on sampling, which impacts phenotype. We present real‐time monitoring growth from phase contrast microscopy images, respiration using optical sensors for dissolved oxygen....
To evaluate the performance of an artificial intelligence (AI) system (Pegasus, Visulytix Ltd., UK*) at detection diabetic retinopathy (DR) from images captured by a handheld portable fundus camera.A cohort 6404 patients (~80% with diabetes mellitus) was screened for retinal diseases using camera (Pictor Plus, Volk Optical Inc., USA) Mexican Advanced Imaging Laboratory Ocular Research. The were graded DR specialists according to Scottish grading scheme. AI evaluated, retrospectively, in...
We propose a framework for Threat Image Projection (TIP) in cargo transmission X-ray imagery. The method exploits the approximately multiplicative nature of imagery to extract library threat items. These items can then be projected into real cargo. show using experimental data that there is no significant qualitative or quantitative difference between images and TIP images. also describe methods adding realistic variation order robustify Machine Learning (ML) based algorithms trained on TIP....
Background Evidence on the practical application of artificial intelligence (AI)-based diabetic retinopathy (DR) screening is needed. Methods Consented participants were screened for DR using retinal imaging with AI interpretation from March 2021 to June at four diabetes clinics in Rwanda. Additionally, images graded by a UK National Health System-certified image grader. grades based International Classification Diabetic Retinopathy grade 2.0 or higher considered referable. The system was...
We present a new approach for biomass assessment in cell culture using disposable microcentrifuge tube. The specially designed tube is fitted with an upper chamber sample loading and lower 5 µL capillary collection during centrifugation. resulting packed volume (PCV) can be quantitatively expressed as the percentage of total sample. study focused on validation method mammalian lines that are widely used bioprocessing. Using several examples, PCV was shown to more precise, rapid, reproducible...
We present a method for automated car detection in xraytransmission images of freight containers. A random forest classifier was used to classify image sub-windows as "car" and "non-car" based on features such intensity log-intensity, well local structures symmetries encoded by Basic Image Features (BIFs) oriented (oBIFs). The proposed approach validated using dataset stream commerce X-ray images. rate 100% achieved while maintaining false alarm 1.23%. Further reduction rate, potentially at...
The security infrastructure is ill-equipped to detect and deter the smuggling of non-explosive devices that enable terror attacks such as those recently perpetrated in western Europe.The detection so-called "small metallic threats" (SMTs) cargo containers currently relies on statistical risk analysis, intelligence reports, visual inspection X-ray images by officers.The latter very slow unreliable due difficulty task: objects potentially spanning less than 50 pixels have be detected...
The current infrastructure for non-intrusive inspection of cargo containers cannot accommodate exploding com-merce volumes and increasingly stringent regulations. There is a pressing need to develop methods automate parts the workflow, enabling expert operators focus on manageable number high-risk images. To tackle this challenge, we developed modular framework automated X-ray image inspection. Employing state-of-the-art machine learning approaches, including deep learning, demonstrate high...
Purpose: To evaluate Pegasus optical coherence tomography (OCT), a clinical decision support software for the identification of features retinal disease from macula OCT scans, across heterogenous populations involving varying patient demographics, device manufacturers, acquisition sites, and operators. Methods: Five thousand five hundred eighty-eight normal anomalous macular volumes (162,721 B-scans), acquired at independent centers in countries, were processed using software. Results...
Transfection with polyethylenimine (PEI) was evaluated as a method for the generation of recombinant Chinese hamster ovary (CHO DG44) cell lines by direct comparison calcium phosphate-DNA coprecipitation (CaPO4) using both green fluorescent protein (GFP) and monoclonal antibody reporter proteins. Following transfection GFP expression vector, proportion GFP-positive cells determined flow cytometry fourfold higher PEI compared to CaPO4 transfection. However, mean level transient highest...
The capacity of milli and micro litre bioreactors to accelerate process development has been successfully demonstrated in traditional biotechnology. However, for regenerative medicine present smaller scale culture methods cannot cope with the wide range processing variables that need be evaluated. Existing microfabricated devices, which could test different a minimum amount resources (e.g. expensive medium), are typically not designed mind. We novel, autoclavable, scale-down device...
Adherent cell lines are widely used across all fields of biology, including drug discovery, toxicity studies, and regenerative medicine. However, adherent processes often limited by a lack advances in culture systems. While suspension benefit from decades development instrumented bioreactors, cultures typically performed static, noninstrumented flasks well-plates. We previously described microfabricated bioreactor that enables high degree control on the microenvironment cells while remaining...
Existing approaches to automated security image analysis focus on the detection of particular classes threat. However, this mode inspection is ineffectual when dealing with mature threat, for which adversaries have refined effective concealment techniques. Furthermore, these methods may be unable detect potential threats that never been seen before. Therefore, in paper, we investigate an anomaly framework, at X-ray patch-level, based on: (i) representations, and (ii) anomalies relative those...
Over 100 million cargo containers that are declared empty on their manifests transported globally each year. Human operators can confirm if is truly by physical inspection or examination of an X-ray image. However, the huge number means confirmation far from complete. Thus, offer opportunity for criminals to smuggle contraband. We report algorithm automatically detecting loads in transmission images. Detection without generation excessive false positives complicated fabric container,...
Previously, we investigated the use of Convolutional Neural Networks (CNNs) to detect so-called Small Metallic Threats (SMTs) hidden amongst legitimate goods inside a cargo container. We trained CNN from scratch on data produced by Threat Image Projection (TIP) framework that generates images with realistic variation robustify performance. The system achieved 90% detection containers contained single SMT, while raising 6% false positives benign containers. best architecture used raw high...
Phase contrast microscopy (PCM) is routinely used for the inspection of adherent cell cultures in all fields biology and biomedicine. Key decisions experimental protocols are often taken by an operator based on typically qualitative observations. However, automated processing analysis PCM images remain challenging due to low between foreground objects (cells) background as well various imaging artefacts. We propose a trainable pixel-wise segmentation approach whereby image structures...