Prediction of Lung Nodule Progression with an Uncertainty-Aware Hierarchical Probabilistic Network.

Diagnostics (Basel)

BCN MedTech, Department of Information and Communication Technologies, Universitat Pompeu Fabra, 08108 Barcelona, Spain.

Published: October 2022

Predicting whether a lung nodule will grow, remain stable or regress over time, especially early in its follow-up, would help doctors prescribe personalized treatments and better surgical planning. However, the multifactorial nature of lung tumour progression hampers the identification of growth patterns. In this work, we propose a deep hierarchical generative and probabilistic network that, given an initial image of the nodule, predicts whether it will grow, quantifies its future size and provides its expected semantic appearance at a future time. Unlike previous solutions, our approach also estimates the uncertainty in the predictions from the intrinsic noise in medical images and the inter-observer variability in the annotations. The evaluation of this method on an independent test set reported a future tumour growth size mean absolute error of 1.74 mm, a nodule segmentation Dice's coefficient of 78% and a tumour growth accuracy of 84% on predictions made up to 24 months ahead. Due to the lack of similar methods for providing future lung tumour growth predictions, along with their associated uncertainty, we adapted equivalent deterministic and alternative generative networks (i.e., probabilistic U-Net, Bayesian test dropout and Pix2Pix). Our method outperformed all these methods, corroborating the adequacy of our approach.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9689366PMC
http://dx.doi.org/10.3390/diagnostics12112639DOI Listing

Publication Analysis

Top Keywords

tumour growth
12
lung nodule
8
probabilistic network
8
will grow
8
lung tumour
8
prediction lung
4
nodule
4
nodule progression
4
progression uncertainty-aware
4
uncertainty-aware hierarchical
4

Similar Publications

Pediatric Sublingual Dermoid Cyst: Diagnostic Challenges and Surgical Management in a 13-Year-Old Boy.

Am J Case Rep

December 2024

Department of Pediatric Otolaryngology, Head and Neck Surgery, Chair of Pediatric Surgery, Medical University of Silesia (SUM), Katowice, Poland.

BACKGROUND Ranulas are typical causes of sublingual cysts in children. However, our case was histopathologically confirmed to be a dermoid cyst. Epidermoid and dermoid cysts of the floor of the mouth account for <0.

View Article and Find Full Text PDF

Inhibition of microRNA-660-5p decreases breast cancer progression through direct targeting of TMEM41B.

Hereditas

December 2024

Facultad de Ciencias Biológicas, Departamento de Biología Celular y Genética, Universidad Autónoma de Nuevo León, San Nicolás de los Garza, Nuevo León, México.

Background: Breast cancer is the most prevalent cancer among women worldwide. Most breast cancer-related deaths result from metastasis and drug resistance. Novel therapies are imperative for targeting metastatic and drug-resistant breast cancer cells.

View Article and Find Full Text PDF

Background: Peritoneal dissemination of ovarian cancer (OvCa) can be largely attributed to the formation of a metastatic microenvironment driven by tumoral exosomes. Here, we aimed to elucidate the mechanisms through which exosomal annexin A2 (ANXA2) derived from OvCa cells induces an HPMC phenotypic shift in favour of peritoneal metastasis.

Methods: Immunohistochemistry and orthotopic and intraperitoneal OvCa xenograft mouse models were used to clarify the relationship between tumour ANXA2 expression and peritoneal metastasis.

View Article and Find Full Text PDF

The TGFβ signaling pathway is known for its pleiotropic functions in a plethora of biological processes. In melanoma, TGFβ signaling promotes invasiveness and metastasis formation. However, its involvement in the response to therapy is controversial.

View Article and Find Full Text PDF

Glioblastoma tumors are the most common and aggressive adult central nervous system malignancy. Nearly all patients experience disease progression, which significantly contributes to disease mortality. Recently, it has been suggested that recurrent tumors may be characterized by a ferroptosis-prone phenotype with a significant decrease in glutathione peroxidase 4 (GPx4) expression.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!