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

Surgery Surgery

The evolution of facial reanimation techniques.

In American journal of otolaryngology ; h5-index 23.0

This review article provides an updated discussion on evidence-based practices related to the evaluation and management of facial paralysis. Ultimately, the goals of facial reanimation include obtaining facial symmetry at rest, providing corneal protection, restoring smile symmetry and facial movement for functional and aesthetic purposes. The treatment of facial nerve injury is highly individualized, especially given the wide heterogeneity regarding the degree of initial neuronal insult and eventual functional outcome. Recent advancements in facial reanimation techniques have better equipped clinicians to approach challenging patient scenarios with reliable, effective strategies. We discuss how technology such as machine learning software has revolutionized pre- and post-intervention assessments and provide an overview of current controversies including timing of intervention, choice of donor nerve, and management of nonflaccid facial palsy with synkinesis. We highlight novel considerations to mainstay conservative management strategies and examine innovations in modern surgical techniques with a focus on gracilis free muscle transfer. Innervation sources, procedural staging, coaptation patterns, and multi-vector and multi-muscle paddle design are modifications that have significantly evolved over the past decade.

Pan Debbie R, Clark Nicholas W, Chiang Harry, Kahmke Russel R, Phillips Brett T, Barrett Dane M

2023-Mar-01

Facial paralysis, Facial reanimation, Free muscle transfer

Radiology Radiology

A new method for predicting the prognosis of ischemic stroke based vascular structure features and lesion location features.

In Clinical imaging

OBJECTIVE : Determining the changes in the prognosis of the cerebral infarction area has an important guiding role in the selection of the treatment plan. The goal of this study is to propose a machine learning-based method that can predict the prognosis of stroke effectively and efficiently.

METHODS : 97 cases of stroke were analyzed retrospectively. Firstly, we extracted vascular structural features from computed tomography angiography (CTA) images and stroke location features from diffusion-weighted imaging (DWI) images to comprehensively characterize the lesions, respectively. Then, we performed sparse representation-based feature selection and classification to predict the prognosis of stroke based on the extracted features. Finally, we randomly divided the 97 cases into cross-validation set, independent testing set 1 and independent testing set 2 to validate the proposed model.

RESULTS : 464 vascular structure features and 116 positional features were extracted. After feature selection, 52 features were finally applied to build the classification model. The proposed model achieved promising prediction performance on the two independent testing sets, with the classification accuracies of 85.19% and 81.25%, respectively.

CONCLUSION : The proposed machine learning approach can effectively mine and accurately quantify the features related to the prognosis, which include the vascular structural features and the stroke location features. In addition, the established prognostic prediction model based on these features has achieved interesting performances, which may provide valuable guidance for the clinical treatment of stroke.

Weng Suiqing, Sun Xilin, Wang Hao, Song Bin, Zhu Jie

2023-Mar-15

Machine learning, Prognosis prediction, Stroke, Stroke location, Vascular structure

General General

Characterization of micropollutants in urban stormwater using high-resolution monitoring and machine learning.

In Water research

Urban rainfall events can lead to the runoff of pollutants, including industrial, pesticide, and pharmaceutical chemicals. Transporting micropollutants (MPs) into water systems can harm both human health and aquatic species. Therefore, it is necessary to investigate the dynamics of MPs during rainfall events. However, few studies have examined MPs during rainfall events due to the high analytical expenses and extensive spatiotemporal variability. Few studies have investigated the occurrence patterns of MPs and factors that influence their transport, such as rainfall duration, antecedent dry periods, and variations in streamflow. Moreover, while there have been many analyses of nutrients, suspended solids, and heavy metals during the first flush effect (FFE), studies on the transport of MPs during FFE are insufficient. This study aimed to identify the dynamics of MPs and FFE in an urban catchment, using high-resolution monitoring and machine learning methods. Hierarchical clustering analysis and partial least squares regression (PLSR) were implemented to estimate the similarity between each MP and identify the factors influencing their transport during rainfall events. Eleven dominant MPs comprised 75% of the total MP concentration and had a 100% detection frequency. During rainfall events, pesticides and pharmaceutical MPs showed a higher FFE than industrial MPs. Moreover, the initial 30% of the runoff volume contained 78.0% of pesticide and 50.1% of pharmaceutical substances for events W1 (July 5 to July 6, 2021) and W6 (August 31 to September 1, 2021), respectively. The PLSR model suggested that stormflow (m3/s) and the duration of antecedent dry hours (h) significantly influenced MP dynamics, yielding the variable importance on projection scores greater than 1.0. Hence, our findings indicate that MPs in urban waters should be managed by considering FFE.

Yun Daeun, Kang Daeho, Cho Kyung Hwa, Baek Sang-Soo, Jeon Junho

2023-Mar-12

First flush effects, Hierarchical clustering analysis, High-resolution mass spectrometry, Micropollutant, Partial least squares regression

General General

Two-Branch network for brain tumor segmentation using attention mechanism and super-resolution reconstruction.

In Computers in biology and medicine

Accurate segmentation of brain tumor plays an important role in MRI diagnosis and treatment monitoring of brain tumor. However, the degree of lesions in each patient's brain tumor region is usually inconsistent, with large structural differences, and brain tumor MR images are characterized by low contrast and blur, current deep learning algorithms often cannot achieve accurate segmentation. To address this problem, we propose a novel end-to-end brain tumor segmentation algorithm by integrating the improved 3D U-Net network and super-resolution image reconstruction into one framework. In addition, the coordinate attention module is embedded before the upsampling operation of the backbone network, which enhances the capture ability of local texture feature information and global location feature information. To demonstrate the segmentation results of the proposed algorithm in different brain tumor MR images, we have trained and evaluated the proposed algorithm on BraTS datasets, and compared with other deep learning algorithms by dice similarity scores. On the BraTS2021 dataset, the proposed algorithm achieves the dice similarity score of 89.61%, 88.30%, 91.05%, and the Hausdorff distance (95%) of 1.414 mm, 7.810 mm, 4.583 mm for the enhancing tumors, tumor cores and whole tumors, respectively. The experimental results illuminate that our method outperforms the baseline 3D U-Net method and yields good performance on different datasets. It indicated that it is robust to segmentation of brain tumor MR images with structures vary considerably.

Jia Zhaohong, Zhu Hongxin, Zhu Junan, Ma Ping

2023-Mar-15

3D U-Net, Attention mechanism, Brain tumor segmentation, MRI, Shared encoder, Super-resolution image reconstruction

General General

PRA-Net: Part-and-Relation Attention Network for depression recognition from facial expression.

In Computers in biology and medicine

Artificial intelligence methods are widely applied to depression recognition and provide an objective solution. Many effective automated methods for detecting depression use facial expressions, which are strong indicators to reflect psychiatric disorders. However, these methods suffer from insufficient representations of depression. To this end, we propose a novel Part-and-Relation Attention Network (PRA-Net), which can enhance depression representations by accurately focusing on features that are highly correlated with depression. Specifically, we first perform partition on the feature map instead of the original image, in order to obtain part features rich in semantic information. Afterwards, self-attention is used to calculate the weight of each part feature. Following, the relationship between the part feature and the global content representation is explored by relation attention to refine the weight. Finally, all features are aggregated into a more compact and depression-informative representation via both weights for depression score prediction. Extensive experiments demonstrate the superiority of our method. Compared to other end-to-end methods, our method achieves state-of-the-art performance on AVEC2013 and AVEC2014.

Liu Zhenyu, Yuan Xiaoyan, Li Yutong, Shangguan Zixuan, Zhou Li, Hu Bin

2023-Jan-24

Automatic depression detection, End-to-end network, Facial expression, Part features, Relation attention

General General

Measuring water holding capacity in pork meat images using deep learning.

In Meat science

Water holding capacity (WHC) plays an important role when obtaining a high-quality pork meat. This attribute is usually estimated by pressing the meat and measuring the amount of water expelled by the sample and absorbed by a filter paper. In this work, we used the Deep Learning (DL) architecture named U-Net to estimate water holding capacity (WHC) from filter paper images of pork samples obtained using the press method. We evaluated the ability of the U-Net to segment the different regions of the WHC images and, since the images are much larger than the traditional input size of the U-Net, we also evaluated its performance when we change the input size. Results show that U-Net can be used to segment the external and internal areas of the WHC images with great precision, even though the difference in the appearance of these areas is subtle.

de Sousa Reis Vinicius Clemente, Ferreira Isaura Maria, Durval Mariah Castro, Antunes Robson Carlos, Backes Andre Ricardo

2023-Mar-16

Computer vision system, Semantic image segmentation, Water holding capacity