RMAU-Net: Residual Multi-Scale Attention U-Net For liver and tumor segmentation in CT images

Jiang, Linfeng; Ou, Jiajie; Liu, Ruihua; Zou, Yangyang; Xie, Ting; Xiao, Hanguang; Bai, Ting

doi:10.1016/j.compbiomed.2023.106838

Cited by 21 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, accurately segmenting small and low-contrast target tissues is a challenge for current AI techniques , explicitly involving the automatic segmentation of sHCCs on plain CT images. U-Net-based deep learning frameworks have demonstrated robustness and scalability in medical image segmentation with extensive research focusing on their application in the automated segmentation of liver tumors. − Our study indicated that by using charged Au NP contrast the 3D U-Net and 3D Trans U-Net models achieved high segmentation accuracy on CT images of tumor-bearing mice. Notably, the 3D Trans U-Net model outperformed the enhanced 3D U-Net model with a mean DSC of 83.50%, as described elsewhere. − HCC shows high contrast and resolution on MR images, which often allows for better automatic segmentation .…”

mentioning

confidence: 85%

“…U-Net-based deep learning frameworks have demonstrated robustness and scalability in medical image segmentation with extensive research focusing on their application in the automated segmentation of liver tumors. − Our study indicated that by using charged Au NP contrast the 3D U-Net and 3D Trans U-Net models achieved high segmentation accuracy on CT images of tumor-bearing mice. Notably, the 3D Trans U-Net model outperformed the enhanced 3D U-Net model with a mean DSC of 83.50%, as described elsewhere. − HCC shows high contrast and resolution on MR images, which often allows for better automatic segmentation . In this study, the automatic segmentation results of sHCC in CT images are consistent with the automatic segmentation results of MR images in previous studies .…”

mentioning

confidence: 85%

See 1 more Smart Citation

Charged Gold Nanoparticles for Target Identification–Alignment and Automatic Segmentation of CT Image-Guided Adaptive Radiotherapy in Small Hepatocellular Carcinoma

Lai,

Luo,

Liu

et al. 2024

Nano Lett.

View full text Add to dashboard Cite

Because of the challenges posed by anatomical uncertainties and the low resolution of plain computed tomography (CT) scans, implementing adaptive radiotherapy (ART) for small hepatocellular carcinoma (sHCC) using artificial intelligence (AI) faces obstacles in tumor identification−alignment and automatic segmentation. The current study aims to improve sHCC imaging for ART using a gold nanoparticle (Au NP)-based CT contrast agent to enhance AI-driven automated image processing. The synthesized charged Au NPs demonstrated notable in vitro aggregation, low cytotoxicity, and minimal organ toxicity. Over time, an in situ sHCC mouse model was established for in vivo CT imaging at multiple time points. The enhanced CT images processed using 3D U-Net and 3D Trans U-Net AI models demonstrated high geometric and dosimetric accuracy. Therefore, charged Au NPs enable accurate and automatic sHCC segmentation in CT images using classical AI models, potentially addressing the technical challenges related to tumor identification, alignment, and automatic segmentation in CT-guided online ART.

show abstract

mentioning

confidence: 85%

mentioning

confidence: 85%

Charged Gold Nanoparticles for Target Identification–Alignment and Automatic Segmentation of CT Image-Guided Adaptive Radiotherapy in Small Hepatocellular Carcinoma

Lai,

Luo,

Liu

et al. 2024

Nano Lett.

View full text Add to dashboard Cite

show abstract

“…Based on the presence of mutations, the authors defined low and high fibroblasts activation classes and, on this variable, they analyzed HCC sensitivity to chemotherapy. Their results showed that patients with high activation of cancer-associated fibroblasts had increased response to chemotherapy.It is also worth to say that many authors have recently published different DL models to perform segmentation of liver cancer and vacularization with the aim to develop a tool useful for physicians to plan liver cancer treatment(21)(22)(23)(24)(25)(26)(27)(28)(29)(30). However, their results are limited to a comparison of accuracy between the proposed model and manual segmentation or segmentation performed by other models, without any proposed practical applications in clinical setting.…”

mentioning

confidence: 99%

Different applications of Artificial Intelligence in liver cancer: a scoping review – part II, from treatment planning and efficacy assessment to prognosis prediction and follow-up. (Preprint)

Chierici,

Lareyre,

Salucki

et al. 2024

Preprint

View full text Add to dashboard Cite

BACKGROUND Artificial Intelligence (AI) plays a pivotal role in early detection and personalized treatment of liver cancer. The integration of AI in screening and diagnosis enhances detection accuracy and aids in formulating effective treatment strategies, but it can be an effective tool to guide liver cancer management in all the steps from treatment deliverance to follow-up. OBJECTIVE This is the second part of a scoping review on implementation of AI and liver cancer, focusing on treatment planning and efficacy assessment, prognosis prediction, and follow-up. METHODS A systematic review was performed on PubMed, Embase, Scopus, and Web of Science databases including research published between January the 1st 2020 and September the 30th 2023. RESULTS AI-driven tools offer predictive analytics for prognosis, treatment planning, and efficacy assessment, aiming to optimize patient outcomes. In liver cancer management, AI assists in treatment planning, such as liver resection and radioembolization, by improving preoperative mapping and predicting therapeutic response. Additionally, AI models predict chemotherapy efficacy based on patient-specific factors, facilitating tailored treatment approaches. Moreover, leveraging AI models, integrating clinical, biochemical, radiological, and histological data, enables accurate prognostication at diagnosis and post-treatment. Key factors such as microvascular invasion, tumor capsule integrity, and grade significantly influence liver cancer prognosis, often assessed using AI-driven predictive models. Imaging modalities, coupled with AI algorithms, exhibit high accuracy in predicting microvascular invasion, aiding treatment planning and prognosis assessment. Following treatment, AI plays a crucial role in prognosis assessment. For patients undergoing liver resection, machine learning models predict disease-free survival, aiding decisions regarding adjuvant chemotherapy. Similarly, models for thermoablation and liver transplantation provide insights into recurrence risk, guiding post-treatment follow-up. In patients receiving systemic treatment like immunotherapy, AI-based models predict cancer-related mortality and overall survival, facilitating treatment response assessment and patient stratification. CONCLUSIONS AI finds wide application in the management of liver cancer treatment and follow-up. Despite promising advancements, challenges remain, including the need for external validation and adaptation to diverse patient populations. Further research is essential to realize the full potential of AI in liver cancer management and translate it into clinical impact. CLINICALTRIAL

show abstract

Literature survey on deep learning methods for liver segmentation from CT images: a comprehensive review

S. S.,

R. S.

2024

Multimed Tools Appl

View full text Add to dashboard Cite

RMAU-Net: Residual Multi-Scale Attention U-Net For liver and tumor segmentation in CT images

Cited by 21 publications

References 31 publications

Charged Gold Nanoparticles for Target Identification–Alignment and Automatic Segmentation of CT Image-Guided Adaptive Radiotherapy in Small Hepatocellular Carcinoma

Charged Gold Nanoparticles for Target Identification–Alignment and Automatic Segmentation of CT Image-Guided Adaptive Radiotherapy in Small Hepatocellular Carcinoma

Different applications of Artificial Intelligence in liver cancer: a scoping review – part II, from treatment planning and efficacy assessment to prognosis prediction and follow-up. (Preprint)

Literature survey on deep learning methods for liver segmentation from CT images: a comprehensive review

Contact Info

Product

Resources

About