A transfer learning approach on MRI-based radiomics signature for overall survival prediction of low-grade and high-grade gliomas

Le, Viet Huan; Minh, Tran Nguyen Tuan; Kha, Quang Hien; Le, Nguyen Quoc Khanh

doi:10.1007/s11517-023-02875-2

Cited by 10 publications

(9 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the United States, the 5-year survival rate for patients with metastatic cancer is only 5%, which is significantly lower than the overall 5-year survival rate of 77%. Therefore, it is very urgent to explore the promising biomarkers for the better prediction of bladder cancer and for seeking diagnosis and treatment for the bladder cancer patients [ 18 ].With the development of multiple images, cancer can be diagnosed easily [ 19 ]. A CT scan is a frequently used technique for diagnosing bladder cancer [ 20 ].…”

Section: Discussionmentioning

confidence: 99%

Analysis of Bladder Cancer Staging Prediction Using Deep Residual Neural Network, Radiomics, and RNA-Seq from High-Definition CT Images

Zhou,

Zheng,

Sun

et al. 2024

Genetics Research

View full text Add to dashboard Cite

Bladder cancer has recently seen an alarming increase in global diagnoses, ascending as a predominant cause of cancer-related mortalities. Given this pressing scenario, there is a burgeoning need to identify effective biomarkers for both the diagnosis and therapeutic guidance of bladder cancer. This study focuses on evaluating the potential of high-definition computed tomography (CT) imagery coupled with RNA-sequencing analysis to accurately predict bladder tumor stages, utilizing deep residual networks. Data for this study, including CT images and RNA-Seq datasets for 82 high-grade bladder cancer patients, were sourced from the TCIA and TCGA databases. We employed Cox and lasso regression analyses to determine radiomics and gene signatures, leading to the identification of a three-factor radiomics signature and a four-gene signature in our bladder cancer cohort. ROC curve analyses underscored the strong predictive capacities of both these signatures. Furthermore, we formulated a nomogram integrating clinical features, radiomics, and gene signatures. This nomogram’s AUC scores stood at 0.870, 0.873, and 0.971 for 1-year, 3-year, and 5-year predictions, respectively. Our model, leveraging radiomics and gene signatures, presents significant promise for enhancing diagnostic precision in bladder cancer prognosis, advocating for its clinical adoption.

show abstract

Section: Discussionmentioning

confidence: 99%

Analysis of Bladder Cancer Staging Prediction Using Deep Residual Neural Network, Radiomics, and RNA-Seq from High-Definition CT Images

Zhou,

Zheng,

Sun

et al. 2024

Genetics Research

View full text Add to dashboard Cite

show abstract

“…Gliomas stand as the most prevalent form of primary malignancy affecting the central nervous system, 7 and represent the leading cause of tumor-induced epilepsy. 8 The prognosis of high-grade gliomas is worse than that of low-grade gliomas, 9 , 10 but the incidence of epilepsy is lower than that of low-grade gliomas, about 40% to 64%. 11 , 12 Gliomas serve as the predominant pathological etiology underlying epilepsy, manifesting in motor and speech impairments that markedly reduce the quality of life of patients.…”

Section: Discussionmentioning

confidence: 99%

Efficacy of Personalized Postoperative Epilepsy Management in Patients with Glioblastoma Utilizing IDH1 Gene Assessment

Meng,

Chen,

et al. 2024

NDT

View full text Add to dashboard Cite

Objective We explored the correlation between the presence of isocitrate dehydrogenase-1 (IDH1) mutations and the incidence of postoperative epilepsy in patients with glioblastoma, as well as assessed the efficacy of preemptive administration of antiepileptic medications in mitigating the occurrence of postoperative epilepsy. Methods Fifty-three patients who received a postoperative pathological diagnosis of glioblastoma, were enrolled in this study. Tumor specimens were subjected to IDH1 gene analysis. The patient cohort was stratified based on their IDH1 mutation status and the administration of prophylactic antiepileptic drugs during the postoperative phase. We subsequently conducted a comparative analysis of postoperative epileptic complications within each patient subgroup. Results In the cohort of 53 patients under study, the occurrence of epilepsy was observed in 10 out of 21 patients carrying IDH1 mutations, while 5 out of 32 patients with wild-type IDH1 also experienced epilepsy, revealing a statistically significant difference (P < 0.05). Among the 27 patients who received prophylactic antiepileptic drugs, 6 of them developed epilepsy, whereas 9 out of 26 patients who did not receive prophylactic antiepileptic drugs exhibited concurrent epilepsy, with no statistically significant difference (P > 0.05). However, when performing a subgroup analysis, it was found that 3 out of 12 patients with IDH1 mutations who received prophylactic antiepileptic drugs experienced epilepsy, whereas 7 out of 9 patients who did not receive prophylactic antiepileptic drugs developed epilepsy, demonstrating a statistically significant difference (P < 0.05). Furthermore, within the group of 15 patients with wild-type IDH1, 3 patients who received prophylactic antiepileptic drugs developed epilepsy, while 2 cases of epilepsy occurred among the 17 patients who did not receive prophylactic antiepileptic drugs, with no statistically significant difference (P > 0.05). Conclusion In individuals with IDH1 mutant glioblastoma who have undergone surgical resection, the implementation of preventive antiepileptic therapy demonstrates a potential to diminish the occurrence of postoperative epilepsy.

show abstract

“…Statistical analyses on zebrafish and mice data were performed with GraphPad Prism 9.3.1 software. Broadly, the statistical analyses follow the statistical analyses that we and others have applied ( Kim, 2015 ; Keefe et al, 2020 ; Witkamp et al, 2022 ; Le et al, 2023 ; Tran et al, 2023 ; McKnight and Najab, n.d. ). Importantly, in certain experiments, WT animals were included to ensure that the disease phenotype in mutant animals was as expected (“the experiment worked”—controls).…”

Section: Discussionmentioning

confidence: 99%

Lithium: effects in animal models of vanishing white matter are not promising

Witkamp,

Oudejans,

Hoogterp

et al. 2024

Front. Neurosci.

View full text Add to dashboard Cite

Vanishing white matter (VWM) is a devastating autosomal recessive leukodystrophy, resulting in neurological deterioration and premature death, and without curative treatment. Pathogenic hypomorphic variants in subunits of the eukaryotic initiation factor 2B (eIF2B) cause VWM. eIF2B is required for regulating the integrated stress response (ISR), a physiological response to cellular stress. In patients’ central nervous system, reduced eIF2B activity causes deregulation of the ISR. In VWM mouse models, the extent of ISR deregulation correlates with disease severity. One approach to restoring eIF2B activity is by inhibition of GSK3β, a kinase that phosphorylates eIF2B and reduces its activity. Lithium, an inhibitor of GSK3β, is thus expected to stimulate eIF2B activity and ameliorate VWM symptoms. The effects of lithium were tested in zebrafish and mouse VWM models. Lithium improved motor behavior in homozygous eif2b5 mutant zebrafish. In lithium-treated 2b4he2b5ho mutant mice, a paradoxical increase in some ISR transcripts was found. Furthermore, at the dosage tested, lithium induced significant polydipsia in both healthy controls and 2b4he2b5ho mutant mice and did not increase the expression of other markers of lithium efficacy. In conclusion, lithium is not a drug of choice for further development in VWM based on the limited or lack of efficacy and significant side-effect profile.

show abstract

A transfer learning approach on MRI-based radiomics signature for overall survival prediction of low-grade and high-grade gliomas

Cited by 10 publications

References 48 publications

Analysis of Bladder Cancer Staging Prediction Using Deep Residual Neural Network, Radiomics, and RNA-Seq from High-Definition CT Images

Analysis of Bladder Cancer Staging Prediction Using Deep Residual Neural Network, Radiomics, and RNA-Seq from High-Definition CT Images

Efficacy of Personalized Postoperative Epilepsy Management in Patients with Glioblastoma Utilizing IDH1 Gene Assessment

Lithium: effects in animal models of vanishing white matter are not promising

Contact Info

Product

Resources

About