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

Dermatology Dermatology

Computer-aided classification of indirect immunofluorescence patterns on esophagus and split skin for the detection of autoimmune dermatoses.

In Frontiers in immunology ; h5-index 100.0

Autoimmune bullous dermatoses (AIBD) are rare diseases that affect human skin and mucous membranes. Clinically, they are characterized by blister formation and/or erosions. Depending on the structures involved and the depth of blister formation, they are grouped into pemphigus diseases, pemphigoid diseases, and dermatitis herpetiformis. Classification of AIBD into their sub-entities is crucial to guide treatment decisions. One of the most sensitive screening methods for initial differentiation of AIBD is the indirect immunofluorescence (IIF) microscopy on tissue sections of monkey esophagus and primate salt-split skin, which are used to detect disease-specific autoantibodies. Interpretation of IIF patterns requires a detailed examination of the image by trained professionals automating this process is a challenging task with these highly complex tissue substrates, but offers the great advantage of an objective result. Here, we present computer-aided classification of esophagus and salt-split skin IIF images. We show how deep networks can be adapted to the specifics and challenges of IIF image analysis by incorporating segmentation of relevant regions into the prediction process, and demonstrate their high accuracy. Using this semi-automatic extension can reduce the workload of professionals when reading tissue sections in IIF testing. Furthermore, these results on highly complex tissue sections show that further integration of semi-automated workflows into the daily workflow of diagnostic laboratories is promising.

Hocke Jens, Krauth Jens, Krause Christopher, Gerlach Stefan, Warnemünde Nicole, Affeldt Kai, van Beek Nina, Schmidt Enno, Voigt Jörn

2023

autoimmune dermatoses, deep learning, immunofluorescence tests, neural networks, tissue classification

General General

Pretreatment with Eupatilin Attenuates Inflammation and Coagulation in Sepsis by Suppressing JAK2/STAT3 Signaling Pathway.

In Journal of inflammation research

PURPOSE : Sepsis is an aggressive and life-threatening organ dysfunction induced by infection. Excessive inflammation and coagulation contribute to the negative outcomes for sepsis, resulting in high morbidity and mortality. In this study, we explored whether Eupatilin could alleviate lung injury, reduce inflammation and coagulation during sepsis.

METHODS : We constructed an in vitro sepsis model by stimulating RAW264.7 cells with 1 μg/mL lipopolysaccharide (LPS) for 6 hours. The cells were divided into control group, LPS group, LPS+ Eupatilin (Eup) group, and Eup group to detect their cell activity and inflammatory cytokines and coagulation factor levels. Cells in LPS+Eup and Eup group were pretreated with Eupatilin (10μM) for 2 hours. In vivo, mice were divided into sham operation group, cecal ligation and puncture (CLP) group and Eup group. Mice in the CLP and Eup groups were pretreated with Eupatilin (10mg/kg) for 2 hours by gavage. Lung tissue and plasma were collected and inflammatory cytokines, coagulation factors and signaling were measured.

RESULTS : In vitro, tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, and tissue factor (TF) expression in LPS-stimulated RAW264.7 cells was downregulated by Eupatilin (10μM). Furthermore, Eupatilin inhibited phosphorylation of the JAK2/STAT3 signaling pathway and suppressed p-STAT3 nuclear translocation. In vivo, Eupatilin increased the survival rate of the mice. In septic mice, plasma concentrations of TNF-α, IL-1β and IL-6, as well as TF, plasminogen activator inhibitor 1 (PAI-1), D-dimer, thrombin-antithrombin complex (TAT) and fibrinogen were improved by Eupatilin. Moreover, Eupatilin alleviated lung injury by improving the expression of inflammatory cytokines and TF, fibrin deposition and macrophage infiltration in lung tissue.

CONCLUSION : Our results revealed that Eupatilin may modulate inflammation and coagulation indicators as well as improve lung injury in sepsis via the JAK2/STAT3 signaling pathway.

Lu Yilun, Li Ding, Huang Yueyue, Sun Yuanyuan, Zhou Hongmin, Ye Fanrong, Yang Hongjing, Xu Tingting, Quan Shichao, Pan Jingye

2023

Eupatilin, JAK2/STAT3, coagulation, inflammation, sepsis

General General

Biological function simulation in neuromorphic devices: from synapse and neuron to behavior.

In Science and technology of advanced materials

As the boom of data storage and processing, brain-inspired computing provides an effective approach to solve the current problem. Various emerging materials and devices have been reported to promote the development of neuromorphic computing. Thereinto, the neuromorphic device represented by memristor has attracted extensive research due to its outstanding property to emulate the brain's functions from synaptic plasticity, sensory-memory neurons to some intelligent behaviors of living creatures. Herein, we mainly review the progress of these brain functions mimicked by neuromorphic devices, concentrating on synapse (i.e. various synaptic plasticity trigger by electricity and/or light), neurons (including the various sensory nervous system) and intelligent behaviors (such as conditioned reflex represented by Pavlov's dog experiment). Finally, some challenges and prospects related to neuromorphic devices are presented.

Chen Hui, Li Huilin, Ma Ting, Han Shuangshuang, Zhao Qiuping

2023

Neuromorphic computing, artificial intelligence, memristor, neuron, synapse

Public Health Public Health

The use of deep learning for smartphone-based human activity recognition.

In Frontiers in public health

The emerging field of digital phenotyping leverages the numerous sensors embedded in a smartphone to better understand its user's current psychological state and behavior, enabling improved health support systems for patients. As part of this work, a common task is to use the smartphone accelerometer to automatically recognize or classify the behavior of the user, known as human activity recognition (HAR). In this article, we present a deep learning method using the Resnet architecture to implement HAR using the popular UniMiB-SHAR public dataset, containing 11,771 measurement segments from 30 users ranging in age between 18 and 60 years. We present a unified deep learning approach based on a Resnet architecture that consistently exceeds the state-of-the-art accuracy and F1-score across all classification tasks and evaluation methods mentioned in the literature. The most notable increase we disclose regards the leave-one-subject-out evaluation, known as the most rigorous evaluation method, where we push the state-of-the-art accuracy from 78.24 to 80.09% and the F1-score from 78.40 to 79.36%. For such results, we resorted to deep learning techniques, such as hyper-parameter tuning, label smoothing, and dropout, which helped regularize the Resnet training and reduced overfitting. We discuss how our approach could easily be adapted to perform HAR in real-time and discuss future research directions.

Stampfler Tristan, Elgendi Mohamed, Fletcher Richard Ribon, Menon Carlo

2023

activity recognition, data science, deep learning, digital health, physical activity, public health, smartphone, wearable technology

General General

Global and non-Global slow oscillations differentiate in their depth profiles.

In Frontiers in network physiology

Sleep slow oscillations (SOs, 0.5-1.5 Hz) are thought to organize activity across cortical and subcortical structures, leading to selective synaptic changes that mediate consolidation of recent memories. Currently, the specific mechanism that allows for this selectively coherent activation across brain regions is not understood. Our previous research has shown that SOs can be classified on the scalp as Global, Local or Frontal, where Global SOs are found in most electrodes within a short time delay and gate long-range information flow during NREM sleep. The functional significance of space-time profiles of SOs hinges on testing if these differential SOs scalp profiles are mirrored by differential depth structure of SOs in the brain. In this study, we built an analytical framework to allow for the characterization of SO depth profiles in space-time across cortical and sub-cortical regions. To test if the two SO types could be differentiated in their cortical-subcortical activity, we trained 30 machine learning classification algorithms to distinguish Global and non-Global SOs within each individual, and repeated this analysis for light (Stage 2, S2) and deep (slow wave sleep, SWS) NREM stages separately. Multiple algorithms reached high performance across all participants, in particular algorithms based on k-nearest neighbors classification principles. Univariate feature ranking and selection showed that the most differentiating features for Global vs. non-Global SOs appeared around the trough of the SO, and in regions including cortex, thalamus, caudate nucleus, and brainstem. Results also indicated that differentiation during S2 required an extended network of current from cortical-subcortical regions, including all regions found in SWS and other basal ganglia regions, and amygdala and hippocampus, suggesting a potential functional differentiation in the role of Global SOs in S2 vs. SWS. We interpret our results as supporting the potential functional difference of Global and non-Global SOs in sleep dynamics.

Seok Sang-Cheol, McDevitt Elizabeth, Mednick Sara C, Malerba Paola

2022

machine learning, sleep, sleep EEG, slow oscillations, space-time

General General

Differentiating acute from chronic insomnia with machine learning from actigraphy time series data.

In Frontiers in network physiology

Acute and chronic insomnia have different causes and may require different treatments. They are investigated with multi-night nocturnal actigraphy data from two sleep studies. Two different wrist-worn actigraphy devices were used to measure physical activities. This required data pre-processing and transformations to smooth the differences between devices. Statistical, power spectrum, fractal and entropy analyses were used to derive features from the actigraphy data. Sleep parameters were also extracted from the signals. The features were then submitted to four machine learning algorithms. The best performing model was able to distinguish acute from chronic insomnia with an accuracy of 81%. The algorithms were then used to evaluate the acute and chronic groups compared to healthy sleepers. The differences between acute insomnia and healthy sleep were more prominent than between chronic insomnia and healthy sleep. This may be associated with the adaptation of the physiology to prolonged periods of disturbed sleep for individuals with chronic insomnia. The new model is a powerful addition to our suite of machine learning models aiming to pre-screen insomnia at home with wearable devices.

Rani S, Shelyag S, Karmakar C, Zhu Ye, Fossion R, Ellis J G, Drummond S P A, Angelova M

2022

actigraphy, acute insomnia, chronic insomnia, dynamical features, insomnia detection, machine learning, sleep parameters