Match!

End-to-end semantic segmentation of personalized deep brain structures for non-invasive brain stimulation.

Published on Feb 25, 2020in Neural Networks5.785
· DOI :10.1016/J.NEUNET.2020.02.006
Essam A. Rashed7
Estimated H-index: 7
(Suez Canal University),
Jose Gomez-Tames8
Estimated H-index: 8
(Nagoya Institute of Technology),
Akimasa Hirata34
Estimated H-index: 34
(Nagoya Institute of Technology)
Abstract
Electro-stimulation or modulation of deep brain regions is commonly used in clinical procedures for the treatment of several nervous system disorders. In particular, transcranial direct current stimulation (tDCS) is widely used as an affordable clinical application that is applied through electrodes attached to the scalp. However, it is difficult to determine the amount and distribution of the electric field (EF) in the different brain regions due to anatomical complexity and high inter-subject variability. Personalized tDCS is an emerging clinical procedure that is used to tolerate electrode montage for accurate targeting. This procedure is guided by computational head models generated from anatomical images such as MRI. Distribution of the EF in segmented head models can be calculated through simulation studies. Therefore, fast, accurate, and feasible segmentation of different brain structures would lead to a better adjustment for customized tDCS studies. In this study, a single-encoder multi-decoders convolutional neural network is proposed for deep brain segmentation. The proposed architecture is trained to segment seven deep brain structures using T1-weighted MRI. Network generated models are compared with a reference model constructed using a semi-automatic method, and it presents a high matching especially in Thalamus (Dice Coefficient (DC) = 94.70%), Caudate (DC = 91.98%) and Putamen (DC = 90.31%) structures. Electric field distribution during tDCS in generated and reference models matched well each other, suggesting its potential usefulness in clinical practice.
  • References (42)
  • Citations (1)
📖 Papers frequently viewed together
20205.81NeuroImage
6 Authors (Oula Puonti, ..., Axel Thielscher)
2 Citations
2015NER: International IEEE/EMBS Conference on Neural Engineering
7 Citations
78% of Scinapse members use related papers. After signing in, all features are FREE.
References42
Newest
#1Jose Gomez-Tames (Nagoya Institute of Technology)H-Index: 8
#2Akihiro Asai (Nagoya Institute of Technology)H-Index: 2
Last. Akimasa Hirata (Nagoya Institute of Technology)H-Index: 34
view all 3 authors...
Abstract Objective Transcranial direct current stimulation (tDCS) is a neuromodulation scheme that delivers a small current via electrodes placed on the scalp. The target is generally assumed to be under the electrode, but deep brain regions could also be involved due to the large current spread between the electrodes. This study aims to computationally evaluate if group-level hotspots exist in deep brain regions for different electrode montages. Methods We computed the tDCS-generated electric f...
5 CitationsSource
#1Nadieh Khalili (UU: Utrecht University)H-Index: 3
#2Nikolas Lessmann (UU: Utrecht University)H-Index: 8
Last. Ivana Išgum (UU: Utrecht University)H-Index: 34
view all 9 authors...
Abstract MR images of fetuses allow clinicians to detect brain abnormalities in an early stage of development. The cornerstone of volumetric and morphologic analysis in fetal MRI is segmentation of the fetal brain into different tissue classes. Manual segmentation is cumbersome and time consuming, hence automatic segmentation could substantially simplify the procedure. However, automatic brain tissue segmentation in these scans is challenging owing to artifacts including intensity inhomogeneity,...
8 CitationsSource
#1Kanghyun Ryu (Yonsei University)H-Index: 3
#2Na-Young Shin (Catholic University of Korea)H-Index: 6
Last. Yoonho Nam (Catholic University of Korea)H-Index: 10
view all 4 authors...
Abstract For quantitative neuroimaging studies using multi-echo gradient echo (mGRE) images, additional T1-weighted magnetization prepared rapid gradient echo (MPRAGE) images are often acquired to supplement the insufficient morphometric information of mGRE for tissue segmentation which require lengthened scan time and additional processing such as image registration. This study investigated the feasibility of generating synthetic MPRAGE images from mGRE images using a deep convolutional neural ...
3 CitationsSource
#1Essam A. Rashed (Suez Canal University)H-Index: 7
#2Jose Gomez-Tames (Nagoya Institute of Technology)H-Index: 8
Last. Akimasa Hirata (Nagoya Institute of Technology)H-Index: 34
view all 3 authors...
Abstract The development of personalized human head models from medical images has become an important topic in the electromagnetic dosimetry field, including the optimization of electrostimulation, safety assessments, etc. Human head models are commonly generated via the segmentation of magnetic resonance images into different anatomical tissues. This process is time consuming and requires special experience for segmenting a relatively large number of tissues. Thus, it is challenging to accurat...
7 CitationsSource
#1Amod Jog (Harvard University)H-Index: 15
#1Amod Jog (Harvard University)H-Index: 2
Last. Bruce Fischl (Harvard University)H-Index: 97
view all 5 authors...
Abstract With the advent of convolutional neural networks (CNN), supervised learning methods are increasingly being used for whole brain segmentation. However, a large, manually annotated training dataset of labeled brain images required to train such supervised methods is frequently difficult to obtain or create. In addition, existing training datasets are generally acquired with a homogeneous magnetic resonance imaging (MRI) acquisition protocol. CNNs trained on such datasets are unable to gen...
7 CitationsSource
Sep 18, 2019 in ICIP (International Conference on Image Processing)
#1Essam A. Rashed (Nagoya Institute of Technology)H-Index: 7
#2Essam A. RashedH-Index: 1
Last. Akimasa Hirata (Nagoya Institute of Technology)H-Index: 34
view all 3 authors...
Transcranial magnetic stimulation (TMS) is a non-invasive clinical technique used for treatment of several neurological diseases such as depression, Alzheimer’s disease and Parkinson’s disease. However, it is always challenging to accurately adjust the electric field on different specific brain regions due to the requirement of several stimulation parameters’ optimizations. A major factor of brain induced electric field is the inter-subject variability, therefore a computer simulation is frequen...
4 CitationsSource
#1Daria Antonenko (Humboldt University of Berlin)H-Index: 16
#2Axel Thielscher (DTU: Technical University of Denmark)H-Index: 33
Last. Agnes Flöel (Humboldt University of Berlin)H-Index: 7
view all 7 authors...
Abstract Background Recent research on neural and behavioral consequences of transcranial direct current stimulation (tDCS) has highlighted the impact of individual factors, such as brain anatomy which determines current field distribution and may thus significantly impact stimulation effects. Computational modeling approaches may significantly advance our understanding of such factors, but the association of simulation-based tDCS-induced fields and neurophysiological outcomes has not been inves...
17 CitationsSource
#1Essam A. Rashed (British University in Egypt)H-Index: 7
#2Takashi Sakai (Nagoya Institute of Technology)H-Index: 1
Last. Akimasa Hirata (Nagoya Institute of Technology)H-Index: 34
view all 4 authors...
Notice to readers This article is available in both Japanese and Spanish languages on the IEEE Pulse website: https://pulse.embs.org .
7 CitationsSource
#1Jose Dolz (École de technologie supérieure)H-Index: 13
#2Karthik Gopinath (École de technologie supérieure)H-Index: 6
Last. Ismail Ben Ayed (École de technologie supérieure)H-Index: 26
view all 6 authors...
Recently, dense connections have attracted substantial attention in computer vision because they facilitate gradient flow and implicit deep supervision during training. Particularly, DenseNet that connects each layer to every other layer in a feed-forward fashion and has shown impressive performances in natural image classification tasks. We propose HyperDenseNet , a 3-D fully convolutional neural network that extends the definition of dense connectivity to multi-modal segmentation problems. Eac...
61 CitationsSource
#1Jose Bernal (University of Girona)H-Index: 6
#2Kaisar Kushibar (University of Girona)H-Index: 5
Last. Xavier Lladó Bardera (University of Girona)H-Index: 2
view all 7 authors...
Abstract In recent years, deep convolutional neural networks (CNNs) have shown record-shattering performance in a variety of computer vision problems, such as visual object recognition, detection and segmentation. These methods have also been utilised in medical image analysis domain for lesion segmentation, anatomical segmentation and classification. We present an extensive literature review of CNN techniques applied in brain magnetic resonance imaging (MRI) analysis, focusing on the architectu...
50 CitationsSource
Cited By1
Newest
#1Aline de Faria Lemos (BME: Budapest University of Technology and Economics)
Last. Balázs Nagy (BME: Budapest University of Technology and Economics)H-Index: 12
view all 2 authors...
In this paper, a method to access the location of a steel strip in the rolling process was developed. The method consists of a hybrid system composed of a CNN-based semantic segmentation followed by morphological operation and outlier removal. The proposed method was capable of estimating the position of the strip with high precision and low computational burden, making it suitable for the application. The implementation of automatic estimation for the steel strip positioning, replacing the curr...
Source