Predicting short-term survival after total resection in glioblastomas by machine learning-based radiomic analysis of preoperative MRI

Radiomics, in combination with artificial intelligence, has emerged as a powerful tool for the development of predictive models in neuro-oncology. Our study aims to find an answer to a clinically relevant question: is there a radiomic profile that can identify glioblastoma (GBM) patients with short-term survival after complete tumor resection? A retrospective study of GBM patients who underwent surgery was conducted in two institutions between January 2019 and January 2020, along with cases from public databases. Cases with gross total or near total tumor resection were included. Preoperative structural multiparametric magnetic resonance imaging (mpMRI) sequences were pre-processed, and a total of 15,720 radiomic features were extracted. After feature reduction, machine learning-based classifiers were used to predict early mortality (<6 months). Additionally, a survival analysis was performed using the random survival forest (RSF) algorithm. A total of 203 patients were enrolled in this study. In the classification task, the naive Bayes classifier obtained the best results in the test data set, with an area under the curve (AUC) of 0.769 and classification accuracy of 80%. The RSF model allowed the stratification of patients into low- and high-risk groups. In the test data set, this model obtained values of C-Index = 0.61, IBS = 0.123 and integrated AUC at six months of 0.761. In this study, we developed a reliable predictive model of short-term survival in GBM by applying open-source and user-friendly computational means. These new tools will assist clinicians in adapting our therapeutic approach considering individual patient characteristics.

Download Full-text

Predicting short-term survival after total resection in glioblastomas by machine learning-based radiomic analysis of preoperative MRI

10.21203/rs.3.rs-640526/v3 ◽

2021 ◽

Author(s):

Santiago Cepeda ◽

Angel Perez-Nuñez ◽

Sergio Garcia-Garcia ◽

Daniel Garcia-Perez ◽

Ignacio Arrese ◽

...

Keyword(s):

Machine Learning ◽

Tumor Resection ◽

Area Under The Curve ◽

Patient Characteristics ◽

Risk Groups ◽

Feature Reduction ◽

Short Term ◽

Term Survival ◽

Reduction Machine ◽

Short Term Survival

Abstract Background Radiomics, in combination with artificial intelligence, has emerged as a powerful tool for the development of predictive models in neuro-oncology. Our study aims to find an answer to a clinically relevant question: is there a radiomic profile that can identify glioblastoma (GBM) patients with short-term survival after complete tumor resection?Methods A retrospective study of GBM patients who underwent surgery was conducted in two institutions between January 2019 and January 2020, along with cases from public databases. Cases with gross total or near-total tumor resection were included. Preoperative structural multiparametric magnetic resonance imaging (mpMRI) sequences were preprocessed, and a total of 15720 radiomic features were extracted. After feature reduction, machine learning-based classifiers were used to predict early mortality (< 6 months). Additionally, a survival analysis was performed using the random survival forest (RSF) algorithm.Results A total of 203 patients were enrolled in this study. In the classification task, the naive Bayes classifier obtained the best results in the testing cohort, with an area under the curve (AUC) of 0.769 and classification accuracy of 80%. The RSF model allowed the stratification of patients into low- and high-risk groups. In the validation set, this model obtained values of C-Index = 0.61, IBS = 0.123 and integrated AUC at six months of 0.761.Conclusion In this study, we developed a reliable predictive model of short-term survival in GBM by applying open-source and user-friendly computational means. These new tools will assist clinicians in adapting our therapeutic approach considering individual patient characteristics.

Download Full-text

Predicting short-term survival after total resection in glioblastomas by machine learning-based radiomic analysis of preoperative MRI

10.21203/rs.3.rs-640526/v1 ◽

2021 ◽

Author(s):

Santiago Cepeda ◽

Angel Perez-Nuñez ◽

Sergio Garcia-Garcia ◽

Daniel Garcia-Perez ◽

Ignacio Arrese ◽

...

Keyword(s):

Machine Learning ◽

Tumor Resection ◽

Area Under The Curve ◽

Patient Characteristics ◽

Risk Groups ◽

Feature Reduction ◽

Short Term ◽

Term Survival ◽

The Individual ◽

Short Term Survival

Abstract Background Radiomics, in combination with artificial intelligence, emerges as a powerful tool for the development of predictive models in neuro-oncology. Our study aims to find an answer to a clinically relevant question: is there a radiomic profile that can identify glioblastoma (GBM) patients with short-term survival after complete tumor resection?Methods A retrospective study of GBM patients who underwent surgery was conducted in two institutions between January 2019 and January 2020, along with cases from public databases. Cases with gross total or near-total tumor resection were included. Preoperative structural multiparametric magnetic resonance imaging (mpMRI) sequences were pre-processed, and a total of 15720 radiomic features were extracted. After feature reduction, machine learning-based classifiers were used to predict early mortality (< 6 months). Also, a survival analysis was performed using the Random Survival Forest (RSF) algorithm. Results A total of 203 patients were enrolled in this study. In the classification task, the Naive Bayes classifier obtained the best results in the testing cohort, with an area under the curve (AUC) of 0.769 and classification accuracy of 80%. RSF model allowed the stratification of patients into low and high-risk groups. In the validation set, this model obtained values of C-Index = 0.61, IBS = 0.123 and integrated AUC at six months of 0.761.Conclusion In this study, we have developed a reliable predictive model of short-term survival in GBM applying open-source and user-friendly computational means. These new tools will assist clinicians in adapting our therapeutic approach considering the individual patient characteristics.

Download Full-text

The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastoma

Cukurova Medical Journal ◽

10.17826/cumj.904688 ◽

2021 ◽

Vol 46 (2) ◽

pp. 706-713

Author(s):

Okan DILEK ◽

Emin DEMIREL ◽

Emre BİLGİN ◽

Berna BOZKURT DUMAN ◽

Bozkurt GÜLEK

Keyword(s):

Machine Learning ◽

Learning Algorithms ◽

Machine Learning Algorithms ◽

Short Term ◽

Term Survival ◽

Enhanced Mri ◽

Contrast Enhanced ◽

Contrast Enhanced Mri ◽

Short Term Survival

Download Full-text

Predicting Short-term Survival after Liver Transplantation using Machine Learning

Scientific Reports ◽

10.1038/s41598-020-62387-z ◽

2020 ◽

Vol 10 (1) ◽

Author(s):

Chien-Liang Liu ◽

Ruey-Shyang Soong ◽

Wei-Chen Lee ◽

Guo-Wei Jiang ◽

Yun-Chun Lin

Keyword(s):

Machine Learning ◽

Liver Transplantation ◽

Short Term ◽

Term Survival ◽

Short Term Survival

Download Full-text

Relationship between the time of day lung transplantation is performed during 24-hours shifts and short-term survival

10.26226/morressier.57108e2fd462b80290b4a68c ◽

2016 ◽

Author(s):

Johannes Menger

Keyword(s):

Lung Transplantation ◽

Time Of Day ◽

Short Term ◽

Term Survival ◽

Short Term Survival

Download Full-text

Osteosarcoma of the jaws: a literature review

Current Medical Imaging Formerly Current Medical Imaging Reviews ◽

10.2174/1573405616666200806173948 ◽

2020 ◽

Vol 16 ◽

Author(s):

Francesco Ricotta ◽

Massimo Bassi ◽

Nicola Tomasetti ◽

Angelo Campobassi ◽

Vincenzo Maiolo ◽

...

Keyword(s):

Treatment Guidelines ◽

State Of The Art ◽

Bone Erosion ◽

Tumor Resection ◽

Bone Destruction ◽

Cell Type ◽

Complete Tumor Resection ◽

Predominant Cell Type ◽

Bone Changes ◽

Complete Tumor

: Osteosarcoma of the jaws (OSJ) is a relatively rare disease, accounting for between 2% and 10% of all cases of osteosarcoma, it is morphologically and radiologically identical to the trunk and extremity variant, but distinct in several crucial aspects. : The lesion is characterized by sarcomatous cells which produces a variable amount of osteoid bone. It arises centrally within the bone and can be subdivided into osteoblastic, chondroblastic and fibroblastic subtype, depending on the predominant cell type. : Radiographically, these tumors display a spectrum of bone changes from well-demarcated borders to lytic bone destruction with indefinite margins and variable cortical bone erosion or, in some cases, images of sclerotic bone. Therapeutic options for OSJ include surgery, chemotherapy and radiotherapy, which are employed according to age of the patient, histological classification and localization of the tumor. Today there is no a general consensus in the treatment guidelines for the OSJ though surgery represents the key of the treatment. The main prognostic factor deeply influencing the patient's prognosis remains the complete tumor resection with negative surgical margins. : The aim of the present review is to describe the state of the art regarding diagnostic and surgical treatment aspects of the primary osteosarcoma of the jaws.

Download Full-text

Influence of age and shock severity on short-term survival in patients with cardiogenic shock

European Heart Journal Acute Cardiovascular Care ◽

10.1093/ehjacc/zuaa035 ◽

2021 ◽

Author(s):

Jacob C Jentzer ◽

Benedikt Schrage ◽

David R Holmes ◽

Salim Dabboura ◽

Nandan S Anavekar ◽

...

Keyword(s):

Cardiogenic Shock ◽

Mayo Clinic ◽

Older Patients ◽

Cox Proportional Hazards ◽

Circulatory Support ◽

Short Term ◽

Term Survival ◽

Risk Of Death ◽

Coronary Syndrome ◽

Short Term Survival

Abstract Aims Cardiogenic shock (CS) is associated with poor outcomes in older patients, but it remains unclear if this is due to higher shock severity. We sought to determine the associations between age and shock severity on mortality among patients with CS. Methods and results Patients with a diagnosis of CS from Mayo Clinic (2007–15) and University Clinic Hamburg (2009–17) were subdivided by age. Shock severity was graded using the Society for Cardiovascular Angiography and Intervention (SCAI) shock stages. Predictors of 30-day survival were determined using Cox proportional-hazards analysis. We included 1749 patients (934 from Mayo Clinic and 815 from University Clinic Hamburg), with a mean age of 67.6 ± 14.6 years, including 33.6% females. Acute coronary syndrome was the cause of CS in 54.0%. The distribution of SCAI shock stages was 24.1%; C, 28.0%; D, 33.2%; and E, 14.8%. Older patients had similar overall shock severity, more co-morbidities, worse kidney function, and decreased use of mechanical circulatory support compared to younger patients. Overall 30-day survival was 53.3% and progressively decreased as age or SCAI shock stage increased, with a clear gradient towards lower 30-day survival as a function of increasing age and SCAI shock stage. Progressively older age groups had incrementally lower adjusted 30-day survival than patients aged <50 years. Conclusion Older patients with CS have lower short-term survival, despite similar shock severity, with a high risk of death in older patients with more severe shock. Further research is needed to determine the optimal treatment strategies for older CS patients.

Download Full-text

Parenteral Nutrition and Oxidant Load in Neonates

Nutrients ◽

10.3390/nu13082631 ◽

2021 ◽

Vol 13 (8) ◽

pp. 2631

Author(s):

Kandeepan Karthigesu ◽

Robert F. Bertolo ◽

Robert J. Brown

Keyword(s):

Parenteral Nutrition ◽

Storage Temperature ◽

Oral Feeding ◽

Published Data ◽

Short Term ◽

Antioxidant Defences ◽

Term Survival ◽

Nutrient Sources ◽

Short Term Survival

Neonates with preterm, gastrointestinal dysfunction and very low birth weights are often intolerant to oral feeding. In such infants, the provision of nutrients via parenteral nutrition (PN) becomes necessary for short-term survival, as well as long-term health. However, the elemental nutrients in PN can be a major source of oxidants due to interactions between nutrients, imbalances of anti- and pro-oxidants, and environmental conditions. Moreover, neonates fed PN are at greater risk of oxidative stress, not only from dietary sources, but also because of immature antioxidant defences. Various interventions can lower the oxidant load in PN, including the supplementation of PN with antioxidant vitamins, glutathione, additional arginine and additional cysteine; reduced levels of pro-oxidant nutrients such as iron; protection from light and oxygen; and proper storage temperature. This narrative review of published data provides insight to oxidant molecules generated in PN, nutrient sources of oxidants, and measures to minimize oxidant levels.

Download Full-text