1
|
Dhiman P, Ma J, Navarro CA, Speich B, Bullock G, Damen JA, Kirtley S, Hooft L, Riley RD, Van Calster B, Moons KGM, Collins GS. Reporting of prognostic clinical prediction models based on machine learning methods in oncology needs to be improved. J Clin Epidemiol 2021; 138:60-72. [PMID: 34214626 PMCID: PMC8592577 DOI: 10.1016/j.jclinepi.2021.06.024] [Citation(s) in RCA: 43] [Impact Index Per Article: 14.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/09/2021] [Revised: 06/15/2021] [Accepted: 06/25/2021] [Indexed: 12/23/2022]
Abstract
OBJECTIVE Evaluate the completeness of reporting of prognostic prediction models developed using machine learning methods in the field of oncology. STUDY DESIGN AND SETTING We conducted a systematic review, searching the MEDLINE and Embase databases between 01/01/2019 and 05/09/2019, for non-imaging studies developing a prognostic clinical prediction model using machine learning methods (as defined by primary study authors) in oncology. We used the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement to assess the reporting quality of included publications. We described overall reporting adherence of included publications and by each section of TRIPOD. RESULTS Sixty-two publications met the inclusion criteria. 48 were development studies and 14 were development with validation studies. 152 models were developed across all publications. Median adherence to TRIPOD reporting items was 41% [range: 10%-67%] and at least 50% adherence was found in 19% (n=12/62) of publications. Adherence was lower in development only studies (median: 38% [range: 10%-67%]); and higher in development with validation studies (median: 49% [range: 33%-59%]). CONCLUSION Reporting of clinical prediction models using machine learning in oncology is poor and needs urgent improvement, so readers and stakeholders can appraise the study methods, understand study findings, and reduce research waste.
Collapse
Affiliation(s)
- Paula Dhiman
- Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, OX3 7LD, UK; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom.
| | - Jie Ma
- Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, OX3 7LD, UK
| | - Constanza Andaur Navarro
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands
| | - Benjamin Speich
- Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, OX3 7LD, UK; Department of Clinical Research, Basel Institute for Clinical Epidemiology and Biostatistics, University Hospital Basel, University of Basel, Basel, Switzerland
| | - Garrett Bullock
- Nuffield Department of Orthopaedics, Rheumatology, and Musculoskeletal Sciences, University of Oxford, Oxford, UK
| | - Johanna Aa Damen
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands
| | - Shona Kirtley
- Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, OX3 7LD, UK
| | - Lotty Hooft
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands
| | - Richard D Riley
- Centre for Prognosis Research, School of Medicine, Keele University, Staffordshire, UK. ST5 5BG
| | - Ben Van Calster
- Department of Development and Regeneration, KU Leuven, Leuven, Belgium.; Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden, the Netherlands.; EPI-centre, KU Leuven, Leuven, Belgium
| | - Karel G M Moons
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands
| | - Gary S Collins
- Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, OX3 7LD, UK; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom
| |
Collapse
|
2
|
Gram IT, Norat T, Rinaldi S, Dossus L, Lukanova A, Téhard B, Clavel-Chapelon F, van Gils CH, van Noord PAH, Peeters PHM, Bueno-de-Mesquita HB, Nagel G, Linseisen J, Lahmann PH, Boeing H, Palli D, Sacerdote C, Panico S, Tumino R, Sieri S, Dorronsoro M, Quirós JR, Navarro CA, Barricarte A, Tormo MJ, González CA, Overvad K, Paaske Johnsen S, Olsen A, Tjønneland A, Travis R, Allen N, Bingham S, Khaw KT, Stattin P, Trichopoulou A, Kalapothaki V, Psaltopoulou T, Casagrande C, Riboli E, Kaaks R. Body mass index, waist circumference and waist–hip ratio and serum levels of IGF-I and IGFBP-3 in European women. Int J Obes (Lond) 2006; 30:1623-31. [PMID: 16552400 DOI: 10.1038/sj.ijo.0803324] [Citation(s) in RCA: 67] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 01/07/2023]
Abstract
OBJECTIVE To examine the relationship between body mass index (BMI) and waist-hip ratio (WHR) with serum levels of insulin-like growth factor-I (IGF-I), and its binding protein (IGFBP)-3. DESIGN Cross-sectional study on 2139 women participating in a case-control study on breast cancer and endogenous hormones. Data on lifestyle and reproductive factors were collected by means of questionnaires. Body height, weight, waist and hip circumferences were measured. Serum levels of IGF-I and insulin-like binding protein (IGFBP)-3 were measured by enzyme-linked immunosorbent assays. Adjusted mean levels of IGF-I and IGFBP-3 across quintiles of BMI, waist circumference, and WHR were calculated by linear regression. Results were adjusted for potential confounders associated with IGF-I and IGFBP-3. RESULTS Adjusted mean serum IGF-I values were lower in women with BMI<22.5 kg/m(2) or BMI>29.2 kg/m(2) compared to women with BMI within this range (P(heterogeneity)<0.0001, P(trend)=0.35). Insulin-like growth factor-I was not related to WHR after adjustment for BMI. IGF-binding protein-3 was linearly positively related to waist and WHR after mutual adjustment. The molar ratio IGF-I/IGFBP-3 had a non-linear relation with BMI and a linear inverse relationship with WHR (P (trend)=0.005). CONCLUSIONS Our data confirm the nonlinear relationship of circulating IGF-I to total adiposity in women. Serum IGFBP-3 was positively related to central adiposity. These suggest that bioavailable IGF-I levels could be lower in obese compared to non-obese women and inversely related to central adiposity.
Collapse
Affiliation(s)
- I T Gram
- Institute of Community Medicine, School of Medicine, University of Tromsø, Tromsø, Norway
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Collapse
|
3
|
Andreoli YE, Laich FS, Navarro CA. [In vitro control of Sclerotinia sclerotiorum and Gaeumannomyces graminis by bacteria of the fluorescent Pseudomonas group]. Rev Argent Microbiol 1993; 25:70-9. [PMID: 8234734] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/29/2023] Open
Abstract
Thirty six fluorescent Pseudomonas isolates were obtained from the rhizosphere of sunflower plants. By antibiosis tests, the six more efficient strains in Sclerotinia sclerotiorum growth inhibition, were selected. Simultaneously, twenty three fluorescent Pseudomonas isolates were recuperated from the rhizosphere of wheat plants and the five most efficient strains in growth inhibition of the fungi Gaeumannomyces graminis were selected. The strains selected from the rhizosphere of sunflower plants had no antagonistic effect on G. graminis and the bacteria isolated from the wheat rhizosphere showed no fungistatic activity on S. sclerotiorum. These results suggest the existence of a certain degree of plant bacteria pathogenic specificity. Among the selected bacteria, the strain FF5 of P. fluorescens originated the major inhibiting halo in vitro against S. sclerotiorum (Figure 1). In liquid culture medium this bacterium produces an antifungal substance that promotes lysis of fungi mycelium (Figure 2) and inhibition of ascospore germination and is not inhibited by the presence of Fe+3 in the culture medium (Table 1). Its synthesis is not associated with the production of fluorescein. Its action is not enzymatic because it is a substance of low molecular weight (< 2000), resistant to autoclave sterilization and photo-stable. The amount of NH4+ and the high pH values produced by the FF5 strain in the liquid culture medium (Table 2) are not responsible for the antifungalal action.
Collapse
Affiliation(s)
- Y E Andreoli
- Facultad de Ciencias Agrarias, Universidad Nacional de Mar del Plata, Argentina
| | | | | |
Collapse
|