TY  - JOUR
AB  - Background: Whole-body imaging has recently been added to large-scale epidemiological studies providing novel opportunities for investigating abdominal organs. However, the segmentation of these organs is required beforehand, which is time consuming, particularly on such a large scale. Methods: We introduce AbdomentNet, a deep neural network for the automated segmentation of abdominal organs on two-point Dixon MRI scans. A pre-processing pipeline enables to process MRI scans from different imaging studies, namely the German National Cohort, UK Biobank, and Kohorte im Raum Augsburg. We chose a total of 61 MRI scans across the three studies for training an ensemble of segmentation networks, which segment eight abdominal organs. Our network presents a novel combination of octave convolutions and squeeze and excitation layers, as well as training with stochastic weight averaging. Results: Our experiments demonstrate that it is beneficial to combine data from different imaging studies to train deep neural networks in contrast to training separate networks. Combining the water and opposed-phase contrasts of the Dixon sequence as input channels, yields the highest segmentation accuracy, compared to single contrast inputs. The mean Dice similarity coefficient is above 0.9 for larger organs liver, spleen, and kidneys, and 0.71 and 0.74 for gallbladder and pancreas, respectively. Conclusions: Our fully automated pipeline provides high-quality segmentations of abdominal organs across population studies. In contrast, a network that is only trained on a single dataset does not generalize well to other datasets.
AU  - Rickmann, A.*
AU  - Senapati, J.*
AU  - Kovalenko, O.*
AU  - Peters, A.
AU  - Bamberg, F.*
AU  - Wachinger, C.*
C1  - 66216
C2  - 52863
TI  - AbdomenNet: Deep neural network for abdominal organ segmentation in epidemiologic imaging studies.
JO  - BMC Med. Imaging
VL  - 22
IS  - 1
PY  - 2022
SN  - 1471-2342
ER  - 

TY  - JOUR
AB  - Background Intraoperative 3-dimensional (3D) navigation is increasingly being used for pedicle screw placement. For this purpose, dedicated mobile 3D C-arms are capable of providing intraoperative fluoroscopy-based 3D image data sets. Modern 3D C-arms have a large field of view, which suggests a higher radiation exposure. In this experimental study we therefore investigate the radiation exposure of a new mobile 3D C-arm with large flat-panel detector to a previously reported device with regular flat-panel detector on an Alderson phantom. Methods We measured the radiation exposure of the Vision RFD 3D (large 30 x 30 cm detector) while creating 3D image sets as well as standard fluoroscopic images of the cervical and lumbar spine using an Alderson phantom. The dosemeter readings were then compared with the radiation exposure of the previous model Vision FD Vario 3D (smaller 20 x 20 cm detector), which had been examined identically in advance and published elsewhere. Results The larger 3D C-arm induced lower radiation exposures at all dosemeter sites in cervical 3D scans as well as at the sites of eye lenses and thyroid gland in lumbar 3D scans. At male and especially female gonads in lumbar 3D scans, however, the larger 3D C-arm showed higher radiation exposures compared with the smaller 3D C-arm. In lumbar fluoroscopic images, the dosemeters near/in the radiation field measured a higher radiation exposure using the larger 3D C-arm. Conclusions The larger 3D C-arm offers the possibility to reduce radiation exposures for specific applications despite its larger flat-panel detector with a larger field of view. However, due to the considerably higher radiation exposure of the larger 3D C-arm during lumbar 3D scans, the smaller 3D C-arm is to be recommended for short-distance instrumentations (mono- and bilevel) from a radiation protection point of view. The larger 3D C-arm with its enlarged 3D image set might be used for long instrumentations of the lumbar spine. From a radiation protection perspective, the use of the respective 3D C-arm should be based on the presented data and the respective application.
AU  - Naseri, Y.*
AU  - Hubbe, U.*
AU  - Scholz, C.*
AU  - Brönner, J.
AU  - Krüger, M.T.*
AU  - Klingler, J.H.*
C1  - 59954
C2  - 49138
CY  - Campus, 4 Crinan St, London N1 9xw, England
TI  - Radiation exposure of a mobile 3D C-arm with large flat-panel detector for intraoperative imaging and navigation-an experimental study using an anthropomorphic Alderson phantom.
JO  - BMC Med. Imaging
VL  - 20
IS  - 1
PB  - Bmc
PY  - 2020
SN  - 1471-2342
ER  -