Lung Nodule Segmentation Using UNet

Kumar, Surender; Bruntha, P. Malin; Daniel, Simon; Kirubakar, J. Ajay; Kiruba, R.; Sam, Siril; Pandian, S. Immanuel Alex

doi:10.1109/icaccs51430.2021.9441977

Cited by 14 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We obtained an effective diameter range between 21.2 and 40.7 cm in thorax exams of the 200 tomographic record samples of the Hospital General de Mexico using the in-house software. Studies in the United States and Chile have reported values of 21.3-33.6 cm and 23-37 cm, respectively [32,33]. In abdomen CT exams, our results showed the effective diameter range was between 22.7 and 33.3 cm; while in the United States, this value has been reported as between 24.8 and 36.5 cm [33], and 21-45 cm in Chile [34].…”

Section: Discussionsupporting

confidence: 40%

See 1 more Smart Citation

CNN application for automated determination of the patient’s size to obtain the size-specific dose estimated in CT

Hernández-Dávila,

Torres-García,

Aranda-Lara

et al. 2024

Biomed. Phys. Eng. Express

View full text Add to dashboard Cite

Introduction. The currently available dosimetry techniques in computed tomography can be inaccurate which overestimate the absorbed dose. Therefore, we aimed to provide an automated and fast methodology to more accurately calculate the SSDE using D_w obtained by using CNN from thorax and abdominal CT study images. Methods. The SSDE was determined from the 200 records files. For that purpose, patients’ size was measured in two ways: a) by developing an algorithm following the AAPM Report No. 204 methodology; and b) using a CNN according to AAPM Report No. 220. Results. The patient’s size measured by the in-house software in the region of thorax and abdomen was 27.63 ± 3.23 cm and 28.66 ± 3.37 cm, while CNN was 18.90 ± 2.6 cm and 21.77 ± 2.45 cm. The SSDE in thorax according to 204 and 220 reports were 17.26 ± 2.81 mGy and 23.70 ± 2.96 mGy for women and 17.08 ± 2.09 mGy and 23.47 ± 2.34 mGy for men. In abdomen was 18.54 ± 2.25 mGy and 23.40 ± 1.88 mGy in women and 18.37 ± 2.31 mGy and 23.84 ± 2.36 mGy in men. Conclusions. Implementing CNN-based automated methodologies can contribute to fast and accurate dose calculations, thereby improving patient-specific radiation safety in clinical practice.

show abstract

Section: Discussionsupporting

confidence: 40%

“…Studies in the United States and Chile have reported values of 21.3-33.6 cm and 23-37 cm, respectively [32,33]. In abdomen CT exams, our results showed the effective diameter range was between 22.7 and 33.3 cm; while in the United States, this value has been reported as between 24.8 and 36.5 cm [33], and 21-45 cm in Chile [34].…”

Section: Discussionsupporting

confidence: 40%

CNN application for automated determination of the patient’s size to obtain the size-specific dose estimated in CT

Hernández-Dávila,

Torres-García,

Aranda-Lara

et al. 2024

Biomed. Phys. Eng. Express

View full text Add to dashboard Cite

show abstract

“…Ref. [20] uses U-Net architecture for lung nodule segmentation for lung cancer detection. In [21], different deep learning models were used for segmenting COVID-19 lung tissues, out of which HRNet achieved high accuracy and dice score.…”

Section: Literature Reviewmentioning

confidence: 99%

“…In [21], different deep learning models were used for segmenting COVID-19 lung tissues, out of which HRNet achieved high accuracy and dice score. The aforementioned previous methods [10,[16][17][18][19][20][21] are used to segment the X-ray chest images using Deep learning, primarily focused on segmenting the image instead of finding lung volume to diagnose the disease.…”

Section: Literature Reviewmentioning

confidence: 99%

Estimating Lung Volume Capacity from X-ray Images Using Deep Learning

Ghimire,

Subedi

2024

QuBS

View full text Add to dashboard Cite

Estimating lung volume capacity is crucial in clinical medicine, especially in disease diagnostics. However, the existing estimation methods are complex and expensive, which require experts to handle and consequently are more error-prone and time-consuming. Thus, developing an automatic measurement system without a human operator that is less prone to human error and, thus, more accurate has always been a prerequisite. The limitation of radiation dose and various medical conditions in technologies like computed tomography was also the primary concern in the past. Although qualitative prediction of lung volume may be a trivial task, designing clinically relevant and automated methods that effectively incorporate imaging data is a challenging problem. This paper proposes a novel multi-tasking-based automatic lung volume estimation method using deep learning that jointly learns segmentation and regression of volume estimation. The two networks, namely, segmentation and regression networks, are sequentially operated with some shared layers. The segmentation network segments the X-ray images, whose output is regressed by the regression network to determine the final lung volume. Besides, the dataset used in the proposed method is collected from three different secondary sources. The experimental results show that the proposed multi-tasking approach performs better than the individual networks. Further analysis of the multi-tasking approach with two different networks, namely, UNet and HRNet, shows that the network with HRNet performs better than the network with UNet with less volume estimation mean square error of 0.0010.

show abstract

“…Recently, deep learning approaches have been proposed for medical image segmentation. The UNet (Ronneberger et al, 2015;Niranjan et al, 2021) is used to capture the background region and obtain accurate localization to identify the ROI. This approach, implemented along with different convolutional layers, provides effective ROI segmentation of input images.…”

Section: Introductionmentioning

confidence: 99%

Lung nodule type classification in CT images using UNet based segmentation and ANFIS based classification

Selvan

2023

CEAI

View full text Add to dashboard Cite

The Early detection and classification of lung nodules using computer-aided diagnosis (CAD) is very much needed to reduce mortality rates in lung cancer patients. The spontaneous and accurate classification of lung nodule type is very supportive for the precise lung cancer diagnosis. In this paper, a novel approach is proposed which employs deep learning for segmentation and ANFIS (Adaptive Neuro Fuzzy Inference System) for lung nodule type classification using CT (Computed Tomography) images. The proposed approach uses UNet architecture for the effective ROI segmentation on the LIDC (Lung Image Database Consortium) database images. A modified ANFIS structure using adjustable parameters to choose the contribution of premise parameters to improve classification accuracy is proposed. The Texture, statistical and shape features of the segmented region is extracted using GLCM (Gray-level Cooccurrence Matrix) algorithm. Then ANFIS classifier is employed to classify the nodules into various types such as juxta-pleural, juxta-vascular, well-circumscribed, pleural-tail and ground-glass opacity. The proposed approach is executed in Python platform and performance of the system is evaluated. The results show that the proposed approach achieves sensitivity, accuracy, precision, specificity and AUC (Area under ROC Curve) of 97.18%, 98.87 %, 98.72%, 98.3% and 0.978 respectively under epoch 30. These results are compared with other lung nodule type classification approaches and found to perform better than other approaches.

show abstract

Lung Nodule Segmentation Using UNet

Cited by 14 publications

References 10 publications

CNN application for automated determination of the patient’s size to obtain the size-specific dose estimated in CT

CNN application for automated determination of the patient’s size to obtain the size-specific dose estimated in CT

Estimating Lung Volume Capacity from X-ray Images Using Deep Learning

Lung nodule type classification in CT images using UNet based segmentation and ANFIS based classification

Contact Info

Product

Resources

About