Narrow Results

With a high rise in deaths caused due to novel coronavirus (nCoV), immunocompromised persons are at high risk. Lung cancer is no exception. Classifying lung cancer patients and Covid-19 is the primary aim of the paper. For this, we propose a deep ensemble neural network (VGG16, DenseNet121, ResNet50 and custom CNN) to detect Covid-19 and lung cancer using chest CT images. We validate our model using three different datasets, namely SPIE AAPM Lung CT Challenge (1503 images), Covid CT dataset (349 images), and SARS-CoV-2 CT-scan dataset (1252 images). We utilize a k(= 5) fold cross-validation approach on the individual deep neural networks (DNNs) and a custom designed CNN model architecture, and achieve a benchmark score of 96.30% (accuracy) with a sensitivity and precision value of 96.39% and 98.44%, respectively. The proposed model effectively utilizes diverse models. To the best of our knowledge, using ensemble DNN, this is the first time we analyze chest CT images to separate lung cancer from Covid-19 (and vice-versa).

As our aim is to classify Covid-19 and lung cancer using chest CT images, it helps in prioritizing immunocompromised persons from Covid-19 for a better patient care. Also, mass screening is possible especially in resource-constrained regions since CT scans are cheaper. The long-term goal is to check whether AI-guided tool(s) is(are) able to prioritize patients that are at high risk (e.g., lung disease) from any possible future infectious disease outbreaks.

INDIA – A novel form of gene regulation in bacteria.

INDIA – Algal biofuels are no energy panacea.

JAPAN – Medical Data Vision enhances the quality of medical care with Actian Vectorwise.

SINGAPORE – Singapore heart surgeon to receive honour from The Royal College of Surgeons of Edinburgh.

SINGAPORE – ELGA^® to deliver innovative water purification at new Singapore General Hospital expansion.

AUSTRALIA – Specialised Therapeutics Australia: New drug to fight hospital superbug infection.

AUSTRALIA – Group of genes hold the clue in migraine cases.

AUSTRALIA – CT scans can triple risk of brain cancer, leukemia.

BRAZIL – Science can do more for sustainable development.

MIDDLE EAST – Particles and persecution: why we should care about Iranian physicists.

EUROPE – Medicyte coordinates EU-funded collaboration on Biomimetic Bioartificial Liver.

EUROPE – Selvita and Orion Pharma achieve a research milestone in Alzheimer's Disease Program.

EUROPE – Zinforo (ceftaroline fosamil) receives positive CHMP opinion in the European Union for the treatment of patients with serious skin infections or community acquired pneumonia.

USA – Vein grown from girl's own stem cells transplanted.

USA – Hidden vitamin in milk yields remarkable health benefits - Weill Cornell researchers show tiny vitamin in milk, in high doses, makes mice leaner, faster and stronger.

USA – New report finds biotechnology companies are participating in 39% of all projects in development for new medicines and technologies for neglected diseases.

USA – TriReme Medical receives FDA clearance for expanded matrix of sizes of Chocolate PTA balloon catheter.

USA – New data show investigational compound dapagliflozin demonstrated significant reductions in blood sugar levels when added to sitagliptin in adults with type 2 diabetes at 24 weeks, with results maintained over 48 weeks.

USA – Zalicus successfully completes Phase 1 single ascending dose study with Z944, a novel, oral T-Type Calcium Channel Blocker.

USA – Study provides clues to clinical trial cost savings.

Finite element (FE) simulations can be utilized to predict contact pressures at the bone/implant interface as well as to identify the position and shape of the contact region. However, the accuracy and reliability of FE models of the bone/implant interface reconstructed from tomographic images may be affected by a number of factors such as the presence of image artifacts, the magnitude of geometric errors made in the reconstruction process, the type of boundary and loading conditions hypothesized in the model, the nonlinear solver utilized for computing the contact pressure distribution, and the element type. This paper attempts to estimate the global effect of the aforementioned factors. For this purpose, a cylindrical contact problem — pin/muff — portraying a simplified model of the bone/implant interface is considered. The accuracy of numerical predictions is estimated by comparing contact pressures predicted by an FE model reconstructed from computed tomography (CT) scan images and by an "ideal", experimentally validated FE model. Two different couplings, i.e. chromium-cobalt alloy and titanium implants, are considered. In the former case, image artifacts complicate the reconstruction process of model geometry and lead to less accurate predictions on contact pressure distribution; conversely, the limited streaking effects occurring in the titanium pin case allow us to precisely reconstruct coupling geometry. Finally, a rather clear correlation between errors on contact pressure and geometric errors made in the reconstruction process is found only for the titanium pin.

Over the last decades, facing the blooming growth of technological progress, interest in digital devices such as computed tomography (CT) as well as magnetic resource imaging which emerged in the 1970s has continued to grow. Such medical data can be invested in numerous visual recognition applications. In this context, these data may be segmented to generate a precise 3D representation of an organ that may be visualized and manipulated to aid surgeons during surgical interventions. Notably, the segmentation process is performed manually through the use of image processing software. Within this framework, multiple outstanding approaches were elaborated. However, the latter proved to be inefficient and required human intervention to opt for the segmentation area appropriately. Over the last few years, automatic methods which are based on deep learning approaches have outperformed the state-of-the-art segmentation approaches due to the use of the relying on Convolutional Neural Networks. In this paper, a segmentation of preoperative patients CT scans based on deep learning architecture was carried out to determine the target organ’s shape. As a result, the segmented 2D CT images are used to generate the patient-specific biomechanical 3D model. To assess the efficiency and reliability of the proposed approach, the 3DIRCADb dataset was invested. The segmentation results were obtained through the implementation of a U-net architecture with good accuracy.

Basically, augmented reality (AR) technology grants innovative ways so as to manipulate and visualize a three-dimensional (3D) model of an object through superimposing computer-generated images onto another object interactively. Being able to interact in real-time with digital and spatial information offers further chances in order to manipulate medical data efficiently and easily. In fact, surgeons encounter multiple challenges handling digital patient data during surgical interventions. Multiple techniques are invested in order to visualize the operative areas, as for example fluoroscopy and ultrasound methods. The latter display certain deficiencies. Therefore, the AR technique can stand for a good candidate in order to project a 3D model of a target organ into the surgeon’s perspective and view field so as to enhance the efficiency and accuracy of the medical intervention intraoperatively. The basic target of this paper is to simulate a generated biomechanical model of the liver organ and visualize its deformations through the use of an AR headset during the open surgery. The proposed approach is validated by the use of acquired CT scans of a human liver organ.

Background: The use of volar locking plate in distal radius fracture can lead to extensor tendon rupture due to dorsal screw penetration. The aim of our study was to investigate the occurrence of dorsal and intra-articular screw penetration using CT scan after volar distal radius osteosynthesis for distal radius fractures.

Methods: Thirty patients who underwent distal volar locking plate for distal radius fracture were included in a retrospective study. In all 30 patients no dorsal and intra-articular screw penetration were detected on standard AP and lateral views of a plain radiograph. CT scan of the operated wrist was performed to determine the number of intra-articular and dorsal screw penetrations. Clinical examination was performed to determine the wrist functions in comparison to the normal wrist.

Results: Nineteen wrists were noted to have screw penetration either dorsally or intraarticularly. The highest incidence is in the 2nd extensor compartment where 13 screws had penetrated with a mean of 2.46 mm. Six screws penetrated into the distal radial ulnar joint and five screws into the wrist joint with a mean of 2.83 mm and 2.6 mm, respectively. However, there was no incidence of tendon irritation or rupture.

Conclusions: This study demonstrated a high incidence of dorsal and intra-articular screw penetration detected by CT scan which was not apparent in plain radiograph. We recommend that surgeons adhere to the principle of only near cortex fixation and downsizing the locking screw length by 2 mm.

Background: Anatomical structure affects function. The morphology of articulations dictates the way forces will travel through the joint. A better understanding of the structure and function of the wrist will enhance our ability to diagnose and treat wrist conditions. Two wrist types have been described based on the morphology of the midcarpal joint. Biomechanically it is important to see if these 2-dimensional (2D) observations reflect articular contact areas. Our purpose was to assess the correlation between measurements performed on wrist radiographs (2D) to measurements performed on 3-dimensional (3D) computed tomography (CT).

Methods: Retrospective review of a database of normal wrist radiographs and corresponding normal CT scans. Only imaging pairs with normal carpal alignment and technically optimal imaging were included. Evaluations included lunate, capitate and wrist type, capitate circumference, percent capitate circumference and volume that articulates with the lunate, scapholunate ligament, scaphoid, hamate, trapezoid, base of the index and middle and ring metacarpal bones.

Results: Midcarpal joint radiographic measurements were positively correlated with measurements on CT scans. Correlations were 0.51 for capitate type and 0.71 for lunate type with both p < 0.001. Percent contact of the lunate with the hamate: r was 0.74 p < 0.001. Using logistic regression analysis, percent lunate-hamate contact on CT was a significant predictor of radiographic lunate type 2 p < 0.001. Percent contact area between lunate and hamate > 7.8% on CT scan achieved a sensitivity of 100% and specificity 79.4% for a type 2 lunate.

Conclusions: 1) Good correlations found between CT and plain radiographs in lunate type, capitate type, and midcarpal joint contact support the use of plain radiographs to describe contact between the carpal bones in the clinical setting. 2) The retrospective nature of this study limited the technical quality of the measurements. Volumetric analysis may aid in a more exact evaluation of surface contact area.

Cancers typically are both highly dangerous and common. Among these, lung cancer has one of the lowest survival rates compared to other cancers. CT scans can reveal dense masses of different shapes and sizes; in the lungs, these are called lung nodules. This study applied a computer-aided diagnosis (CAD) system to detect candidate nodules — and diagnose it either solitary or juxtapleural — with equivalent diameters, ranging from 7.78mm to 22.48mm in a 2D CT slice. Pre-processing and segmentation is a very important step to segment and enhance the CT image. A segmentation and enhancement algorithm is achieved using bilateral filtering, Thresholding the gray-level transformation function, Bounding box and maximum intensity projection. Border artifacts are removed by clearing the lung border, erosion, dilation and superimposing. Feature extraction is done by extracting 20 gray-level co-occurrence matrix features from four directions: $0^{\circ }$, $45^{\circ }$, $90^{\circ }$ and $135^{\circ }$ and one distance of separation ($d=1$ pixel). In the classification step, two classifiers are proposed to classify two types of nodules based on their locations: as juxtapleural or solitary nodules. The two classifiers are a deep learning convolutional neural network (CNN) and the K-nearest neighbor (KNN) algorithm. Random oversampling and 10-fold cross-validation are used to improve the results. In our CAD system, the highest accuracy and sensitivity rates achieved by the CNN were 96% and 95%, respectively, for solitary nodule detection. The highest accuracy and sensitivity rates achieved by the KNN model were 93.8% and 96.7%, respectively, and K was set to 1 to detect juxtapleural nodules.

Image de-noising is an essential tool for removing unwanted signals from an image. In Computed Tomography (CT) images, the image quality is degraded by the absorption of X-rays and quantum noise, which is generated due to the excitement of X-ray photons. Removal of noise and preservation of information in the CT images becomes a challenge for an imaging algorithm design. During the algorithm design selection of dataset is an important aspect for deducing results. The dataset used in this research comprises of 60 CT scan images of liver cancer archived from the arterial contrast enhanced phase. In this phase the cancer cells appear more intense as compared to the healthy liver tissue due to the absorption of contrast enhancing reagent. The experimentation for appropriate noise removal filter selection is done by testing the images using Mean, Median and Weiner Filters. The filter selected should give an image output which has minimal randomness, sharper boundaries and no blur. The de-noised image will provide a better visibility of the disease to the radiologist and physician. The performance parameters used for the assessment of various filters used in the study include visual assessment, entropy and signal to noise ratio (SNR) of the images. Median filter gives an accuracy of 96%, mean filter is 76.2% accurate with respect to original information and Weiner filters has an accuracy of 79.7%.