1
|
Zhao D, Grist JT, Rose HEL, Davies NP, Wilson M, MacPherson L, Abernethy LJ, Avula S, Pizer B, Gutierrez DR, Jaspan T, Morgan PS, Mitra D, Bailey S, Sawlani V, Arvanitis TN, Sun Y, Peet AC. Metabolite selection for machine learning in childhood brain tumour classification. NMR Biomed 2022; 35:e4673. [PMID: 35088473 DOI: 10.1002/nbm.4673] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/04/2021] [Revised: 12/01/2021] [Accepted: 12/02/2021] [Indexed: 06/14/2023]
Abstract
MRS can provide high accuracy in the diagnosis of childhood brain tumours when combined with machine learning. A feature selection method such as principal component analysis is commonly used to reduce the dimensionality of metabolite profiles prior to classification. However, an alternative approach of identifying the optimal set of metabolites has not been fully evaluated, possibly due to the challenges of defining this for a multi-class problem. This study aims to investigate metabolite selection from in vivo MRS for childhood brain tumour classification. Multi-site 1.5 T and 3 T cohorts of patients with a brain tumour and histological diagnosis of ependymoma, medulloblastoma and pilocytic astrocytoma were retrospectively evaluated. Dimensionality reduction was undertaken by selecting metabolite concentrations through multi-class receiver operating characteristics and compared with principal component analysis. Classification accuracy was determined through leave-one-out and k-fold cross-validation. Metabolites identified as crucial in tumour classification include myo-inositol (P < 0.05, AUC = 0 . 81 ± 0 . 01 ), total lipids and macromolecules at 0.9 ppm (P < 0.05, AUC = 0 . 78 ± 0 . 01 ) and total creatine (P < 0.05, AUC = 0 . 77 ± 0 . 01 ) for the 1.5 T cohort, and glycine (P < 0.05, AUC = 0 . 79 ± 0 . 01 ), total N-acetylaspartate (P < 0.05, AUC = 0 . 79 ± 0 . 01 ) and total choline (P < 0.05, AUC = 0 . 75 ± 0 . 01 ) for the 3 T cohort. Compared with the principal components, the selected metabolites were able to provide significantly improved discrimination between the tumours through most classifiers (P < 0.05). The highest balanced classification accuracy determined through leave-one-out cross-validation was 85% for 1.5 T 1 H-MRS through support vector machine and 75% for 3 T 1 H-MRS through linear discriminant analysis after oversampling the minority. The study suggests that a group of crucial metabolites helps to achieve better discrimination between childhood brain tumours.
Collapse
Affiliation(s)
- Dadi Zhao
- Institute of Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK
- Oncology, Birmingham Children's Hospital, Birmingham, UK
| | - James T Grist
- Institute of Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK
- Oncology, Birmingham Children's Hospital, Birmingham, UK
| | - Heather E L Rose
- Institute of Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK
- Oncology, Birmingham Children's Hospital, Birmingham, UK
| | - Nigel P Davies
- Institute of Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK
- Oncology, Birmingham Children's Hospital, Birmingham, UK
- Imaging and Medical Physics, University Hospitals Birmingham, Birmingham, UK
| | - Martin Wilson
- Centre for Human Brain Health, University of Birmingham, Birmingham, UK
| | | | | | | | - Barry Pizer
- Paediatric Oncology, Alder Hey Children's Hospital, Liverpool, UK
| | - Daniel R Gutierrez
- Children's Brain Tumour Research Centre, University of Nottingham, Nottingham, UK
- Medical Physics, Nottingham University Hospitals NHS Trust, Nottingham, UK
| | - Tim Jaspan
- Children's Brain Tumour Research Centre, University of Nottingham, Nottingham, UK
- Neuroradiology, Nottingham University Hospitals NHS Trust, Nottingham, UK
| | - Paul S Morgan
- Children's Brain Tumour Research Centre, University of Nottingham, Nottingham, UK
- Medical Physics, Nottingham University Hospitals NHS Trust, Nottingham, UK
- Division of Clinical Neuroscience, University of Nottingham, Nottingham, UK
| | - Dipayan Mitra
- Neuroradiology, The Newcastle upon Tyne Hospitals, Newcastle upon Tyne, UK
| | - Simon Bailey
- Paediatric Oncology, Great North Children's Hospital, Newcastle upon Tyne, UK
| | - Vijay Sawlani
- Radiology, Queen Elizabeth Hospital Birmingham, Birmingham, UK
- School of Psychology, University of Birmingham, Birmingham, UK
| | - Theodoros N Arvanitis
- Institute of Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK
- Oncology, Birmingham Children's Hospital, Birmingham, UK
- Institute of Digital Healthcare, WMG, University of Warwick, Coventry, UK
| | - Yu Sun
- Institute of Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK
- Oncology, Birmingham Children's Hospital, Birmingham, UK
- University of Birmingham and Southeast University Joint Research Centre for Biomedical Engineering, Suzhou, China
| | - Andrew C Peet
- Institute of Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK
- Oncology, Birmingham Children's Hospital, Birmingham, UK
| |
Collapse
|
2
|
Basquiera AL, García MJ, Martinez Rolón J, Olmedo J, Laviano J, Burgos R, Caeiro G, Remaggi G, Raña P, Paoletti M, González CM, Fernández I, Pavlovsky A, Perusini MA, Rodriguez A, Guanchiale L, Carvani A, Mandrile L, Figueroa F, Vicente Reparaz A, Fragapane Mathus PN, Garate G, Fauque ME, Kantor G, Cruset S, Gonzalez Lorch JS, Szelagowski M, Giarini MP, Oliveira N, García MC, Ventriglia MV, Pereyra PH, Gutierrez DR, Kusminsky G, Troccoli J, Freitas MJ, Cranco S, Del V Sanchez N, Rey I, Funes ME, Jarchum S, Freue J, Miroli A, Guerrero O, López Ares L, Campestri R, Bove V, Salinas GN, Cabrejo M, Milone JH, Zabaljauregui S, Gotta D, Dupont JC, Stemmelin G. Clinical characteristics and evolution of hematological patients and COVID-19 in Argentina: a report from the Argentine Society of Hematology. Medicina (B Aires) 2021; 81:536-545. [PMID: 34453794] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 06/13/2023] Open
Abstract
Individuals with malignancies and COVID-19 have a lower survival compared with the general population. However, the information about the impact of COVID-19 on the whole hematological population is scarce. We aimed to describe the 30th day overall survival (OS) after COVID-19 infection in patients with a hematological disease in Argentina. A completely anonymous survey from the Argentine Society of Hematology was delivered to all the hematologists in Argentina; it started in April 2020. A cut-off to analyze the data was performed in December 2020 and, finally, 419 patients were reported and suitable for the analysis (average age: 58 years, 90% with malignant diseases). After the COVID-19 diagnosis, the 30-day OS for the whole population was 80.2%. From the entire group (419), 101 (24.1%) individuals required intensive care unit admission, where the 30-day OS was 46.6%. Among allogeneic stem cell transplant recipients, the 30-day OS was 70.3%. Factors associated with a low OS were two or more comorbidities, an active hematological disease and history of chemotherapy. In individuals with the three factors, the 30-day OS was 49.6% while the 30-day OS in those without those factors was 100%. Patients with hematological diseases have a higher mortality than the general population. This group represents a challenge and requires careful decision-making of the treatment in order not to compromise the chances of cure.
Collapse
Affiliation(s)
- Ana L Basquiera
- Hospital Privado Universitario de Córdoba, Córdoba, Argentina. E-mail:
| | - Mercedes J García
- Hospital Privado Universitario de Córdoba, Córdoba, Argentina
- Hospital Raúl Ferreyra, Córdoba, Argentina
| | | | | | | | - Rubén Burgos
- Conciencia-Instituto Oncohematológico de la Patagonia, Argentina
| | - Gastón Caeiro
- Hospital Privado Universitario de Córdoba, Córdoba, Argentina
- Hospital Raúl Ferreyra, Córdoba, Argentina
| | | | - Pablo Raña
- Conciencia-Instituto Oncohematológico de la Patagonia, Argentina
| | | | | | | | | | | | | | | | - Analia Carvani
- Hospital Diego Paroissien, Isidro Casanova, Buenos Aires, Argentina
| | - Laura Mandrile
- Hospital Diego Paroissien, Isidro Casanova, Buenos Aires, Argentina
| | | | | | | | | | | | | | - Soledad Cruset
- Hospital El Cruce, Florencio Varela, Buenos Aires, Argentina
| | | | | | | | | | | | | | | | | | | | | | - María J Freitas
- Hospital Nacional Posadas, El Palomar, Buenos Aires, Argentina
| | | | | | - Irene Rey
- Hospital Ramos Mejía, Buenos Aires, Argentina
| | | | | | | | | | | | | | | | - Victor Bove
- Instituto Oncohematológico Privado, Río Cuarto, Córdoba, Argentina
| | | | - María Cabrejo
- Sanatorio Municipal Julio Méndez, Buenos Aires, Argentina
| | | | | | - Daniel Gotta
- Hospital Urquiza, Concepción del Uruguay, Entre Ríos, Argentina
| | | | | |
Collapse
|