Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

Pathology Pathology

Attempt to Predict A/T/N-Based Alzheimer's Disease Cerebrospinal Fluid Biomarkers Using a Peripheral Blood DNA Methylation Clock.

In Journal of Alzheimer's disease reports

Background : Although aging is the strongest risk factor for the development of Alzheimer's disease (AD), it remains uncertain if the blood DNA methylation clock, which reflects the effect of biological aging on DNA methylation (DNAme) status of blood cells, may be used as a surrogate biomarker for AD pathology in the central nervous system (CNS).

Objective : We aimed to develop a practical model to predict for A/T/N-based AD biomarkers as the prediction targets using the aging acceleration of blood cells.

Methods : We obtained data of North American ADNI study participants (n = 317) whose blood DNA methylation microarray (Illumina HumanMethylation EPIC Beadchips) and cerebrospinal fluid (CSF) AD biomarkers (Aβ, t-tau, and p-tau) were recorded simultaneously. Methylation clock was calculated to conduct machine learning, in order to predict binary statuses (+ or -) for A (corresponding to the lowered CSF Aβ), T (the elevated CSF p-tau), or N (the elevated CSF t-tau). The predictive performance of the models was evaluated by area under curve (AUC) in the test subset within ADNI.

Results : Among the 317 included samples, 194 (61.2%) were A+, 247 (77.9%) were T+, and 104 (32.8%) were N+. The degree of blood aging acceleration showed weak positive correlation with the CSF Aβ levels, even after adjustment with APOE genotype and other covariates. However, the contribution of aging acceleration to improve the predictive performance of models was not significant for any of A+, T+, or N+.

Conclusion : Our exploratory attempts could not demonstrate the substantial utility of the peripheral blood cells' methylation clock as a predictor for A/T/N-based CSF biomarkers of AD, and further additional work should be conducted to determine whether the blood DNAme signatures including methylation clock have substantial utility in detecting underlying amyloid, tau or neurodegeneration pathology of AD.

Sato Kenichiro, Mano Tatsuo, Suzuki Kazushi, Toda Tatsushi, Iwatsubo Takeshi, Iwata Atsushi

2020-Jul-23

A/T/N, Alzheimer’s disease, DNA methylation, blood biomarker, epigenetic clock

Surgery Surgery

Adapting for the future: flexibility of UK postgraduate training.

In Surgery (Oxford, Oxfordshire)

Postgraduate surgical training has undergone repeated reforms alongside changes in terms of employment. The broad structure of progression from Foundation years through core and specialist training to the award of a Certificate of Completion of Training is likely to continue for the foreseeable future. Technological developments including robotics, genomics and artificial intelligence together with an extension of the surgical team are likely to alter dramatically the nature of surgery in the future. Surgical training will need to incorporate training in new technologies, including simulation, which will be provided in the workplace, academic institutions and commercial facilities. There will be greater emphasis on non-technical skills and human factors, especially in relation to the use of new technologies and working in wider teams, including non-medical staff. Genomics will play an increasing role in determining individualized patient care, with a need for surgeons to have an understanding of this field and communicate this to their patients. Surgical training will need to be suitably flexible in order to accommodate these developments, to allow more part-time working and portfolio careers, and to encourage recruitment and retention.

Mitchell Tim

2020-Aug-30

Extended surgical team, future of surgery, genomics, non-technical skills, postgraduate medical education, simulation, surgical robots

General General

Low Molecular Weight Volatile Organic Compounds Indicate Grazing by the Marine Rotifer Brachionus plicatilis on the Microalgae Microchloropsis salina.

In Metabolites

Microalgae produce specific chemicals indicative of stress and/or death. The aim of this study was to perform non-destructive monitoring of algal culture systems, in the presence and absence of grazers, to identify potential biomarkers of incipient pond crashes. Here, we report ten volatile organic compounds (VOCs) that are robustly generated by the marine alga, Microchloropsis salina, in the presence and/or absence of the marine grazer, Brachionus plicatilis. We cultured M. salina with and without B. plicatilis and collected in situ volatile headspace samples using thermal desorption tubes over the course of several days. Data from four experiments were aggregated, deconvoluted, and chromatographically aligned to determine VOCs with tentative identifications made via mass spectral library matching. VOCs generated by algae in the presence of actively grazing rotifers were confirmed via pure analytical standards to be pentane, 3-pentanone, 3-methylhexane, and 2-methylfuran. Six other VOCs were less specifically associated with grazing but were still commonly observed between the four replicate experiments. Through this work, we identified four biomarkers of rotifer grazing that indicate algal stress/death. This will aid machine learning algorithms to chemically define and diagnose algal mass production cultures and save algae cultures from imminent crash to make biofuel an alternative energy possibility.

Fisher Carolyn L, Lane Pamela D, Russell Marion, Maddalena Randy, Lane Todd W

2020-Sep-04

Brachionus plicatilis, Microchloropsis salina, biomarkers, headspace sampling, pond crash, volatile organic compounds

General General

Classification of Coronavirus (COVID-19) from X-ray and CT images using shrunken features.

In International journal of imaging systems and technology

Necessary screenings must be performed to control the spread of the COVID-19 in daily life and to make a preliminary diagnosis of suspicious cases. The long duration of pathological laboratory tests and the suspicious test results led the researchers to focus on different fields. Fast and accurate diagnoses are essential for effective interventions for COVID-19. The information obtained by using X-ray and Computed Tomography (CT) images is vital in making clinical diagnoses. Therefore it is aimed to develop a machine learning method for the detection of viral epidemics by analyzing X-ray and CT images. In this study, images belonging to six situations, including coronavirus images, are classified using a two-stage data enhancement approach. Since the number of images in the dataset is deficient and unbalanced, a shallow image augmentation approach was used in the first phase. It is more convenient to analyze these images with hand-crafted feature extraction methods because the dataset newly created is still insufficient to train a deep architecture. Therefore, the Synthetic minority over-sampling technique algorithm is the second data enhancement step of this study. Finally, the feature vector is reduced in size by using a stacked auto-encoder and principal component analysis methods to remove interconnected features in the feature vector. According to the obtained results, it is seen that the proposed method has leveraging performance, especially to make the diagnosis of COVID-19 in a short time and effectively. Also, it is thought to be a source of inspiration for future studies for deficient and unbalanced datasets.

Öztürk Şaban, Özkaya Umut, Barstuğan Mücahid

2020-Aug-18

COVID‐19, classification, coronavirus, feature extraction, hand‐crafted features, sAE

Pathology Pathology

MedMeshCNN -- Enabling MeshCNN for Medical Surface Models

ArXiv Preprint

Background and objective: MeshCNN is a recently proposed Deep Learning framework that drew attention due to its direct operation on irregular, non-uniform 3D meshes. On selected benchmarking datasets, it outperformed state-of-the-art methods within classification and segmentation tasks. Especially, the medical domain provides a large amount of complex 3D surface models that may benefit from processing with MeshCNN. However, several limitations prevent outstanding performances of MeshCNN on highly diverse medical surface models. Within this work, we propose MedMeshCNN as an expansion for complex, diverse, and fine-grained medical data. Methods: MedMeshCNN follows the functionality of MeshCNN with a significantly increased memory efficiency that allows retaining patient-specific properties during the segmentation process. Furthermore, it enables the segmentation of pathological structures that often come with highly imbalanced class distributions. Results: We tested the performance of MedMeshCNN on a complex part segmentation task of intracranial aneurysms and their surrounding vessel structures and reached a mean Intersection over Union of 63.24\%. The pathological aneurysm is segmented with an Intersection over Union of 71.4\%. Conclusions: These results demonstrate that MedMeshCNN enables the application of MeshCNN on complex, fine-grained medical surface meshes. The imbalanced class distribution deriving from the pathological finding is considered by MedMeshCNN and patient-specific properties are mostly retained during the segmentation process.

Lisa Schneider, Annika Niemann, Oliver Beuing, Bernhard Preim, Sylvia Saalfeld

2020-09-10

General General

Conceptualising technology, its development and future: The six genres of technology.

In Technological forecasting and social change

One approach to developing futuristic views of technology is to draw upon experience and expertise. However, this becomes increasingly speculative as one moves to more distant timelines and visionary technological forms. This raises the question of whether it is possible to rationally predict how a technology development trajectory might unfold into the future, perhaps to some 'ultimate form', that is accessible, surfaces the necessary technological features for development as well as considers the implications for human-artefact relationships. The proposed approach is conceptually grounded in a parsimonious framework that examines different configurations of human-artefact relationships, revealing 'Six Genres of Technology'. This suggests how the shift from human-human to artefact-artefact and the increasing autonomy of the artefacts (technological beings), introduces specific features to each of the six Genres. Four features are identified in the later Genres that in combination, could be construed as, or indeed pose a threat: autonomy, intelligence, language, and autopoiesis. This paper advances the debate about future technological developments by using the proposed framework to structure an argument about the key issues that should be discussed today - so that the developments of tomorrow can be more reflectively considered, appropriately debated and knowingly pursued.

Harwood Stephen, Eaves Sally

2020-Nov

Artificial intelligence, Autonomous technology, Digital technology, Digital transformation, Futures, Social impact