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

Ophthalmology Ophthalmology

Development and validation of deep learning algorithms for automated eye laterality detection with anterior segment photography.

In Scientific reports ; h5-index 158.0

This paper aimed to develop and validate a deep learning (DL) model for automated detection of the laterality of the eye on anterior segment photographs. Anterior segment photographs for training a DL model were collected with the Scheimpflug anterior segment analyzer. We applied transfer learning and fine-tuning of pre-trained deep convolutional neural networks (InceptionV3, VGG16, MobileNetV2) to develop DL models for determining the eye laterality. Testing datasets, from Scheimpflug and slit-lamp digital camera photography, were employed to test the DL model, and the results were compared with a classification performed by human experts. The performance of the DL model was evaluated by accuracy, sensitivity, specificity, operating characteristic curves, and corresponding area under the curve values. A total of 14,468 photographs were collected for the development of DL models. After training for 100 epochs, the DL models of the InceptionV3 mode achieved the area under the receiver operating characteristic curve of 0.998 (with 95% CI 0.924-0.958) for detecting eye laterality. In the external testing dataset (76 primary gaze photographs taken by a digital camera), the DL model achieves an accuracy of 96.1% (95% CI 91.7%-100%), which is better than an accuracy of 72.3% (95% CI 62.2%-82.4%), 82.8% (95% CI 78.7%-86.9%) and 86.8% (95% CI 82.5%-91.1%) achieved by human graders. Our study demonstrated that this high-performing DL model can be used for automated labeling for the laterality of eyes. Our DL model is useful for managing a large volume of the anterior segment images with a slit-lamp camera in the clinical setting.

Zheng Ce, Xie Xiaolin, Wang Zhilei, Li Wen, Chen Jili, Qiao Tong, Qian Zhuyun, Liu Hui, Liang Jianheng, Chen Xu

2021-Jan-12

General General

Amorphous metal oxide semiconductor thin film, analog memristor, and autonomous local learning for neuromorphic systems.

In Scientific reports ; h5-index 158.0

Artificial intelligence is a promising concept in modern and future societies. Presently, software programs are used but with a bulky computer size and large power consumption. Conversely, hardware systems named neuromorphic systems are suggested, with a compact computer size and low power consumption. An important factor is the number of processing elements that can be integrated. In the present study, three decisive technologies are proposed: (1) amorphous metal oxide semiconductor thin films, one of which, Ga-Sn-O (GTO) thin film, is used. GTO thin film does not contain rare metals and can be deposited by a simple process at room temperature. Here, oxygen-poor and oxygen-rich layers are stacked. GTO memristors are formed at cross points in a crossbar array; (2) analog memristor, in which, continuous and infinite information can be memorized in a single device. Here, the electrical conductance gradually changes when a voltage is applied to the GTO memristor. This is the effect of the drift and diffusion of the oxygen vacancies (Vo); and (3) autonomous local learning, i.e., extra control circuits are not required since a single device autonomously modifies its own electrical characteristic. Finally, a neuromorphic system is assembled using the abovementioned three technologies. The function of the letter recognition is confirmed, which can be regarded as an associative memory, a typical artificial intelligence application.

Kimura Mutsumi, Sumida Ryo, Kurasaki Ayata, Imai Takahito, Takishita Yuta, Nakashima Yasuhiko

2021-Jan-12

General General

Deep learning-based pupil model predicts time and spectral dependent light responses.

In Scientific reports ; h5-index 158.0

Although research has made significant findings in the neurophysiological process behind the pupillary light reflex, the temporal prediction of the pupil diameter triggered by polychromatic or chromatic stimulus spectra is still not possible. State of the art pupil models rested in estimating a static diameter at the equilibrium-state for spectra along the Planckian locus. Neither the temporal receptor-weighting nor the spectral-dependent adaptation behaviour of the afferent pupil control path is mapped in such functions. Here we propose a deep learning-driven concept of a pupil model, which reconstructs the pupil's time course either from photometric and colourimetric or receptor-based stimulus quantities. By merging feed-forward neural networks with a biomechanical differential equation, we predict the temporal pupil light response with a mean absolute error below 0.1 mm from polychromatic (2007 [Formula: see text] 1 K, 4983 [Formula: see text] 3 K, 10,138 [Formula: see text] 22 K) and chromatic spectra (450 nm, 530 nm, 610 nm, 660 nm) at 100.01 ± 0.25 cd/m2. This non-parametric and self-learning concept could open the door to a generalized description of the pupil behaviour.

Zandi Babak, Khanh Tran Quoc

2021-Jan-12

General General

Sequential transfer learning based on hierarchical clustering for improved performance in deep learning based food segmentation.

In Scientific reports ; h5-index 158.0

Accurately segmenting foods from optical images is a challenging task, yet becoming possible with the help of recent advances in Deep Learning based solutions. Automated identification of food items opens up possibilities of useful applications like nutrition intake monitoring. Given large variations in food choices, Deep Learning based solutions still struggle to generate human level accuracy. In this work, we propose a novel Sequential Transfer Learning method using Hierarchical Clustering. This novel approach simulates a step by step problem solving framework based on clustering of similar types of foods. The proposed approach provides up to 6% gain in accuracy compared to traditional network training and generated a robust model performing better in challenging unseen cases. This approach is also tested for segmenting foods in Danish school children meals for dietary intake monitoring as an application.

Siemon Mia S N, Shihavuddin A S M, Ravn-Haren Gitte

2021-Jan-12

Pathology Pathology

Spontaneous mutations in the single TTN gene represent high tumor mutation burden.

In NPJ genomic medicine

Tumor mutation burden (TMB) is an emerging biomarker, whose calculation requires targeted sequencing of many genes. We investigated if the measurement of mutation counts within a single gene is representative of TMB. Whole-exome sequencing (WES) data from the pan-cancer cohort (n = 10,224) of TCGA, and targeted sequencing (tNGS) and TTN gene sequencing from 24 colorectal cancer samples (AMC cohort) were analyzed. TTN was identified as the most frequently mutated gene within the pan-cancer cohort, and its mutation number best correlated with TMB assessed by WES (rho = 0.917, p < 2.2e-16). Colorectal cancer was one of good candidates for the application of this diagnostic model of TTN-TMB, and the correlation coefficients were 0.936 and 0.92 for TMB by WES and TMB by tNGS, respectively. Higher than expected TTN mutation frequencies observed in other FLAGS (FrequentLy mutAted GeneS) are associated with late replication time. Diagnostic accuracy for high TMB group did not differ between TTN-TMB and TMB assessed by tNGS. Classification modeling by machine learning using TTN-TMB for MSI-H diagnosis was constructed, and the diagnostic accuracy was 0.873 by area under the curve in external validation. TTN mutation was enriched in samples possessing high immunostimulatory signatures. We suggest that the mutation load within TTN represents high TMB status.

Oh Ji-Hye, Jang Se Jin, Kim Jihun, Sohn Insuk, Lee Ji-Young, Cho Eun Jeong, Chun Sung-Min, Sung Chang Ohk

2020-Jan-14

General General

H2AK121ub in Arabidopsis associates with a less accessible chromatin state at transcriptional regulation hotspots.

In Nature communications ; h5-index 260.0

Although it is well established that the Polycomb Group (PcG) complexes maintain gene repression through the incorporation of H2AK121ub and H3K27me3, little is known about the effect of these modifications on chromatin accessibility, which is fundamental to understand PcG function. Here, by integrating chromatin accessibility, histone marks and expression analyses in different Arabidopsis PcG mutants, we show that PcG function regulates chromatin accessibility. We find that H2AK121ub is associated with a less accessible but still permissive chromatin at transcriptional regulation hotspots. Accessibility is further reduced by EMF1 acting in collaboration with PRC2 activity. Consequently, H2AK121ub/H3K27me3 marks are linked to inaccessible although responsive chromatin. In contrast, only-H3K27me3-marked chromatin is less responsive, indicating that H2AK121ub-marked hotspots are required for transcriptional responses. Nevertheless, despite the loss of PcG activities leads to increased chromatin accessibility, this is not necessarily accompanied by transcriptional activation, indicating that accessible chromatin is not always predictive of gene expression.

Yin Xiaochang, Romero-Campero Francisco J, de Los Reyes Pedro, Yan Peng, Yang Jing, Tian Guangmei, Yang XiaoZeng, Mo Xiaorong, Zhao Shuangshuang, Calonje Myriam, Zhou Yue

2021-01-12