Subsequent Disease Diagnosis Research Articles

ResCoWNet: A deep convolutional neural network with residual learning based on DT-CWT for despeckling Optical Coherence Tomography images

Various retinal illnesses can be diagnosed and treated using the non-invasive diagnostic technology called OCT (Optical Coherence Tomography). However, the processes used to create OCT images generate speckle noise which considerably reduces the quality of the images, and the subsequent disease diagnosis is negatively impacted by such poor-quality images. The despeckling is effected using conventional methods such as spatial/transform domain filtering, dictionary learning, or a combination of these methods. Deep Convolutional Neural Networks (CNN) have improved the ability to exploit spatial correlations and extract data at various resolutions by adopting a hierarchical network structure, thereby making image-denoising methods more reliable. In deep networks, the Discrete Wavelet Transform (DWT) and its inverse are typically used in place of pooling operations. Nevertheless, the Dual-Tree Complex Wavelet Transform (DT-CWT) can provide a significant performance enhancement compared to deep network models for noise reduction based on DWT pooling. Hence to despeckle OCT images, this paper proposes a novel deep convolutional neural network (ResCoWNet) with residual learning based on DT-CWT. The input noisy image is transformed using DT-CWT into various subbands and is made to pass through several convolutional as well as residual blocks at three different levels in ResCoWNet. The denoised versions of OCT images produced by our model outperform the experimental results both quantitatively and subjectively.

Risk factors, subsequent disease onset and prognostic impact of myocardial infarction and atrial fibrillation

Abstract Background Myocardial infarction (MI) is a known risk factor for incident atrial fibrillation (AF), while AF frequently complicates acute MI. Although both diseases share common cardiovascular risk factors, the direction and strength of the association of the risk factors with disease onset, subsequent disease incidence and mortality are not completely understood. Purpose Our goal was to define the temporal relationship of MI and AF and the association of cardiovascular risk factors with disease incidence in order to determine whether common clinical risk factors show different associations with incident MI or AF. We further aimed to investigate predictors of subsequent disease onset and the impact of subsequent disease diagnosis on mortality. Methods In pooled multivariable Cox regression analyses we examined temporal relations of disease onset and identified predictors of MI, AF and subsequent all-cause mortality in 108,363 individuals (median age 46.0 years, 48.2% men) free of MI and AF at baseline from six European population-based cohorts. Results Over a maximum follow-up of 10.0 years 3558 (3.3%) individuals were diagnosed exclusively with MI, 1922 (1.8%) with AF but no MI, and 491 (0.5%) individuals developed both MI and AF. Association of male sex, systolic blood pressure, antihypertensive treatment and diabetes mellitus appeared to be stronger with incident MI than with AF, whereas increasing age and body mass index showed a higher risk for incident AF. Total cholesterol and daily smoking were significantly related to incident MI but not AF. The combined population attributable fraction of the cardiovascular risk factors was over 70% for incident MI, whereas it was only about one quarter for incident AF. Subsequent MI after incident AF (hazard ratio1.68, 95% CI 1.03–2.74) and subsequent AF after MI (hazard ratio 1.75, 95% CI 1.31–2.34) both significantly increased overall mortality risk. Conclusions Subsequent diagnosis of MI and AF was associated with a significant increase in mortality, irrespective of the first event. We found different associations of common cardiovascular risk factors with incident MI and AF indicating distinct pathophysiological pathways in disease development. Funding Acknowledgement Type of funding sources: Public grant(s) – EU funding. Main funding source(s): European Union's Horizon 2020 research and innovation programme (grant agreement No 847770, AFFECT-EU) European Union's Horizon 2020 research and innovation programme (grant agreement No 648131)

Image of plant disease segmentation model based on improved pulse-coupled neural network

Image segmentation is a key step in feature extraction and disease recognition of plant diseases images. To avoid subjectivity while using a pulse-coupled neural network (PCNN) which realises parameter configuration through artificial exploration to segment plant disease images, an improved image segmentation model called SFLA-PCNN is proposed in this paper. The shuffled frog-leaping algorithm (SFLA) is used to optimise the parameters (β, αθ, Vθ) of PCNN to improve PCNN performance. A series of plant disease images are taken as segmentation experiments, and the results reveal that SFLA-PCNN is more accurate than other methods mentioned in this paper and can extract lesion images from the background area effectively, providing a foundation for subsequent disease diagnosis.

Fully Convolutional DenseNet with Multiscale Context for Automated Breast Tumor Segmentation.

Breast tumor segmentation plays a crucial role in subsequent disease diagnosis, and most algorithms need interactive prior to firstly locate tumors and perform segmentation based on tumor-centric candidates. In this paper, we propose a fully convolutional network to achieve automatic segmentation of breast tumor in an end-to-end manner. Considering the diversity of shape and size for malignant tumors in the digital mammograms, we introduce multiscale image information into the fully convolutional dense network architecture to improve the segmentation precision. Multiple sampling rates of atrous convolution are concatenated to acquire different field-of-views of image features without adding additional number of parameters to avoid over fitting. Weighted loss function is also employed during training according to the proportion of the tumor pixels in the entire image, in order to weaken unbalanced classes problem. Qualitative and quantitative comparisons demonstrate that the proposed algorithm can achieve automatic tumor segmentation and has high segmentation precision for various size and shapes of tumor images without preprocessing and postprocessing.

Bone Diagenesis and its Implication for Disease Diagnosis: The Relevance of Bone Microstructure Analysis for the Study of Past Human Remains

When bone is exposed to the burial environment it may experience structural changes induced by multiple agents. The study of postmortem alterations is important to differentiate decomposition phenomena from normal physiological processes or pathological lesions, as well as to assess bone tissue quality. Microscopy is of great utility to evaluate the integrity of bone microstructure and it provides significant data on long-term bone decomposition. A total of 18 human bone sections (eight archeological and ten retrieved from an identified skeletal collection) were selected for analysis under plane light and polarized light. The aim of this exploratory study was to analyze the impact of diagenesis and taphonomy on the bone microstructure, as well as on the differential diagnosis of pathological conditions. The results showed that the microscopy approach to bone tissues contributed materially as an aid in the detailed description of the main diagenetic changes observed. It showed that gross inspection does not provide a realistic assessment of bone tissue preservation, which can impact in the characterization of lesions present and subsequent disease diagnosis. Therefore, researchers should continue to consider the application of histological techniques if the aim is to comprehend tissue integrity and its association with decomposition or disease.

Development of a modified selective medium to enhance the recovery rate of Brachyspira hyodysenteriae and other porcine intestinal spirochaetes from faeces

The aim of this study was to develop a modified selective medium to improve the recovery rate of Brachyspira hyodysenteriae and other clinically significant intestinal spirochaetes from porcine faeces. The susceptibility of five Brachyspira spp. type strains and five Thai field isolates of B. hyodysenteriae to the antimicrobials halquinol and flavomycin was determined by in vitro susceptibility tests in the agar dilution method, and optimal incorporation rates were confirmed by broth dilution. All the spirochaetes were susceptible to halquinol at ≤ 1 μg ml(-1), while 16 μg ml(-1) of flavomycin (F) allowed their growth, and therefore, only the latter was selected for further use. F and different combinations of colistin (C), spectinomycin (S) and rifampacin (R) were incorporated into pre-enrichment broths and/or agar plates, and growth of the spirochaetes from seeded faeces was determined. Two solid media were selected for further testing using faeces from 90 finishing pigs on 10 farms. A previously recommended method of pre-enrichment did not increase the recovery rate. The use of blood agar modified medium (BAM) containing F (16 μg ml(-1)), S (400 μg ml(-1)), R (30 μg ml(-1)) and colistin (C, 100 U ml(-1)) (assigning as BAM-CSRF) reduced the growth of contaminating intestinal microbiota and resulted in a significantly higher rate of spirochaete recovery than the previous recommended medium. BAM-CSRF is a useful new selective medium for the isolation of B. hyodysenteriae and other intestinal spirochaetes from pig faeces. The new selective medium for isolating B. hyodysenteriae and other Brachyspira spp. from pig faeces will improve their recovery and subsequent disease diagnosis.