Model Selection Based Algorithm in Neonatal Chest EIT

Seifnaraghi, Nima; Gelidi, Serena de; Nordebo, Sven; Kallio, Merja; Frerichs, Inéz; Tizzard, Andrew; Suo‐Palosaari, Maria; Sophocleous, Louiza; Kaam, Anton H. van; Sorantin, Erich; Demosthenous, Andreas; Bayford, Richard

doi:10.1109/tbme.2021.3053463

Cited by 7 publications

(5 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both the absolute values and the location of the silent spaces were different among three different models in almost every studied subject (table 1). This coincided with the findings that the effect of improper chest model selection may be huge (Seifnaraghi et al 2021) and the information derived from such inappropriately calculated silent spaces may be deceiving. Therefore, the results suggested that the subjectspecific meshes and lung contours may be needed.…”

Section: Discussionsupporting

confidence: 75%

The calculation of electrical impedance tomography based silent spaces requires individual thorax and lung contours

Yang¹,

Fu²,

Frerichs³

et al. 2022

Physiol. Meas.

Self Cite

View full text Add to dashboard Cite

Objective: The present study evaluates the influence of different thorax contours (generic vs individual) on the parameter “silent spaces” computed from electrical impedance tomography (EIT) measurements. Approach: Six patients with acute respiratory distress syndrome were analyzed retrospectively. EIT measurements were performed and the silent spaces were calculated based on (1) patient-specific contours Sind, (2) generic adult male contours SEidorsA and (3) generic neonate contours SEidorsN. Main results: The differences among all studied subjects were 5±6% and 8±7% for Sind vs. SEidorsA, Sind vs. SEidorsN, respectively (median ± interquartile range). Sind values were higher than the generic ones in two patients. Significance: In the present study, we demonstrated the differences in values when the silent spaces were calculated based on different body and organ contours. To our knowledge, this study was the first one showing explicitly that silent spaces calculated with generic thorax and lung contours might lead to results with different locations and values as compared to the calculation with subject-specific models. Interpretations of silent spaces should be proceeded with caution.

show abstract

Section: Discussionsupporting

confidence: 75%

The calculation of electrical impedance tomography based silent spaces requires individual thorax and lung contours

Yang¹,

Fu²,

Frerichs³

et al. 2022

Physiol. Meas.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The fitted trapezoid was further rounded with a radius corresponding to 25% of its height at all corners. The impact of using inaccurate existing models on clinically important parameters in lung monitoring is detailed in [ 11 ].…”

Section: Discussionmentioning

confidence: 99%

Cross-sectional chest circumference and shape development in infants

et al. 2022

Self Cite

View full text Add to dashboard Cite

Objective This study investigates the development of the thoracic cross-section at the nipple line level during the early stages of life. Unlike the descriptive awareness regarding chest development course, there exist no quantitative references concerning shape, circumference and possible dependencies to age, gender or body weight. The proposed mathematical relations are expected to help create guidelines for more realistic modelling and potential detection of abnormalities. One potential application is lung electrical impedance tomography (EIT) monitoring where accurate chest models are crucial in both extracting reliable parameters for regional ventilation function and design of EIT belts. Despite their importance, such reference data is not readily available for the younger age range due to insufficient data amid the regulations of neonatal imaging. Results Chest circumference shows the highest correlation to body weight following the relation $$f\left(x\right)= 18.3735 \ \mathrm{ln}\left(0.0012x+2.1010\right)$$ f x = 18.3735 ln 0.0012 x + 2.1010 where x is the body weight in grams and f(x) is the chest circumference in cm at the nipple line level. No statistically significant difference in chest circumference between genders was detected. However, the shape indicated signs of both age and gender dependencies with on average boys developing a more rectangular shape than girls from the age of 1 years and 9 months.

show abstract

“…As with previous model-based methods [39,40], there are two alternatives for training the network: end-to-end training of the complete system consisting of k max blocks, or sequential training of each block. Nevertheless, for our scenario, conducting end-toend training is not feasible because of two primary factors.…”

Section: Graph Convolutional Neural Networkmentioning

confidence: 99%

A Graph Neural Network Approach with Improved Levenberg–Marquardt for Electrical Impedance Tomography

Zhao,

Xu,

Zhu

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

Electrical impedance tomography (EIT) is a non-invasive imaging method that allows for the acquisition of resistivity distribution information within an object without the use of radiation. EIT is widely used in various fields, such as medical imaging, industrial imaging, geological exploration, etc. Presently, most electrical impedance imaging methods are restricted to uniform domains, such as pixelated pictures. These algorithms rely on model learning-based image reconstruction techniques, which often necessitate interpolation and embedding if the fundamental imaging model is solved on a non-uniform grid. EIT technology still confronts several obstacles today, such as insufficient prior information, severe pathological conditions, numerous imaging artifacts, etc. In this paper, we propose a new electrical impedance tomography algorithm based on the graph convolutional neural network model. Our algorithm transforms the finite-element model (FEM) grid data from the ill-posed problem of EIT into a network graph within the graph convolutional neural network model. Subsequently, the parameters in the non-linear inverse problem of the EIT process are updated by using the improved Levenberg—Marquardt (ILM) method. This method generates an image that reflects the electrical impedance. The experimental results demonstrate the robust generalizability of our proposed algorithm, showcasing its effectiveness across different domain shapes, grids, and non-distributed data.

show abstract

Model Selection Based Algorithm in Neonatal Chest EIT

Cited by 7 publications

References 42 publications

The calculation of electrical impedance tomography based silent spaces requires individual thorax and lung contours

The calculation of electrical impedance tomography based silent spaces requires individual thorax and lung contours

Cross-sectional chest circumference and shape development in infants

A Graph Neural Network Approach with Improved Levenberg–Marquardt for Electrical Impedance Tomography

Contact Info

Product

Resources

About