Predicting outcomes of pelvic exenteration using machine learning
| Title: | Predicting outcomes of pelvic exenteration using machine learning |
|---|---|
| Authors: | Dudurych I.; Kelly M. E.; Aalbers A. G. J.; Abdul Aziz N.; Abecasis N.; Abraham-Nordling M.; Akiyoshi T.; Alberda W.; Albert M.; Andric M.; Angenete E.; Antoniou A.; Auer R.; Austin K. K.; Aziz O.; Baker R. P.; Bali M.; Baseckas G.; Bebington B.; Bedford M.; Bednarski B. K.; Beets G. L.; Berg P. L.; Beynon J.; Biondo S.; Boyle K.; Bordeianou L.; Bremers A. B.; Brunner M.; Buchwald P.; Bui A.; Burgess A.; Burger J. W. A.; Burling D.; Burns E.; Campain N.; Carvalhal S.; Castro L.; Caycedo-Marulanda A.; Chan K. K. L.; Chang G. J.; Chew M. H.; Chok A. K.; Chong P.; Christensen H. K.; Clouston H.; Codd M.; Collins D.; Colquhoun A. J.; Corr A.; Coscia M.; Coyne P. E.; Creavin B.; Croner R. S.; Damjanovic L.; Daniels I. R.; Davies M.; Davies R. J.; Delaney C. P.; Wilt J. H. W.; Denost Q.; Deutsch C.; Dietz D.; Domingo S.; Dozois E. J.; Duff M.; Eglinton T.; Enrique-Navascues J. M.; Espin-Basany E.; Evans M. D.; Fearnhead N. S.; Flatmark K.; Fleming F.; Frizelle F. A.; Gallego M. A.; Garcia-Granero E.; Garcia-Sabrido J. L.; Gentilini L.; George M. L.; George V.; Ghouti L.; Giner F.; Ginther N.; Glynn R.; Golda T.; Griffiths B.; Harris D. A.; Hagemans J. A. W.; Hanchanale V.; Harji D. P.; Helewa R. M.; Heriot A. G.; Hochman D.; Hohenberger W.; Holm T.; Hompes R.; Jenkins J. T.; Kaffenberger S.; Kandaswamy G. V.; Kapur S.; Kanemitsu Y.; Kelley S. R.; Keller D. S.; Khan M. S.; Kiran R. P.; Kim H.; Kim H. J.; Koh C. E.; Kok N. F. M.; Kokelaar R.; Kontovounisios C.; Kristensen H. O.; Kroon H. M.; Kusters M.; Lago V.; Larsen S. G.; Larson D. W.; Law W. L.; Laurberg S.; Lee P. J.; Limbert M.; Lydrup M. L.; Lyons A.; Lynch A. C.; Mantyh C.; Mathis K. L.; Margues C. F. S.; Martling A.; Meijerink W. J. H. J.; Merkel S.; Mehta A. M.; McArthur D. R.; McDermott F. D.; McGrath J. S.; Malde S.; Mirnezami A.; Monson J. R. T.; Morton J. R.; Mullaney T. G.; Negoi I.; Neto J. W. M.; Nguyen B.; Nielsen M. B.; Nieuwenhuijzen G. A. P.; Nilsson P. J.; Oliver A.; O'Connell P. R.; O'Dwyer S. T.; Palmer G.; Pappou E.; Park J.; Patsouras D.; Pellino G.; Peterson A. C.; Poggioli G.; Proud D.; Quinn M.; Quyn A.; Radwan R. W.; Rasheed S.; Rasmussen P. C.; Regenbogen S. E.; Renehan A.; Rocha R.; Rochester M.; Rohila J.; Rothbarth J.; Rottoli M.; Roxburgh C.; Rutten H. J. T.; Ryan E. J.; Safar B.; Sagar P. M.; Sahai A.; Saklani A.; Sammour T.; Sayyed R.; Schizas A. M. P.; Schwarzkopf E.; Scripcariu V.; Selvasekar C.; Shaikh I.; Shellawell G.; Shida D.; Simpson A.; Smart N. J.; Smart P.; Smith J. J.; Solbakken A. M.; Solomon M. J.; Sorensen M. M.; Steele S. R.; Steffens D.; Stitzenberg K.; Stocchi L.; Stylianides N. A.; Swartling T.; Sumrien H.; Sutton P. A.; Swartking T.; Tan E. J.; Taylor C.; Tekkis P. P.; Teras J.; Thurairaja R.; Toh E. L.; Tsarkov P.; Tsukada Y.; Tsukamoto S.; Tuech J. J.; Turner W. H.; Tuynman J. B.; van Ramshorst G. H.; van Zoggel D.; Vasquez-Jimenez W.; Verhoef C.; Vizzielli G.; Voogt E. L. K.; Uehara K.; Wakeman C.; Warrier S.; Wasmuth H. H.; Weber K.; Weiser M. R.; Wheeler J. M. D.; Wild J.; Wilson M.; Wolthuis A.; Yano H.; Yip B.; Yip J.; Yoo R. N.; Winter D. C. |
| Contributors: | Dudurych, I.; Kelly, M. E.; Aalbers, A. G. J.; Abdul Aziz, N.; Abecasis, N.; Abraham-Nordling, M.; Akiyoshi, T.; Alberda, W.; Albert, M.; Andric, M.; Angenete, E.; Antoniou, A.; Auer, R.; Austin, K. K.; Aziz, O.; Baker, R. P.; Bali, M.; Baseckas, G.; Bebington, B.; Bedford, M.; Bednarski, B. K.; Beets, G. L.; Berg, P. L.; Beynon, J.; Biondo, S.; Boyle, K.; Bordeianou, L.; Bremers, A. B.; Brunner, M.; Buchwald, P.; Bui, A.; Burgess, A.; Burger, J. W. A.; Burling, D.; Burns, E.; Campain, N.; Carvalhal, S.; Castro, L.; Caycedo-Marulanda, A.; Chan, K. K. L.; Chang, G. J.; Chew, M. H.; Chok, A. K.; Chong, P.; Christensen, H. K.; Clouston, H.; Codd, M.; Collins, D.; Colquhoun, A. J.; Corr, A.; Coscia, M.; Coyne, P. E.; Creavin, B.; Croner, R. S.; Damjanovic, L.; Daniels, I. R.; Davies, M.; Davies, R. J.; Delaney, C. P.; Wilt, J. H. W.; Denost, Q.; Deutsch, C.; Dietz, D.; Domingo, S.; Dozois, E. J.; Duff, M.; Eglinton, T.; Enrique-Navascues, J. M.; Espin-Basany, E.; Evans, M. D.; Fearnhead, N. S.; Flatmark, K.; Fleming, F.; Frizelle, F. A.; Gallego, M. A.; Garcia-Granero, E.; Garcia-Sabrido, J. L.; Gentilini, L.; George, M. L.; George, V.; Ghouti, L.; Giner, F.; Ginther, N.; Glynn, R.; Golda, T.; Griffiths, B.; Harris, D. A.; Hagemans, J. A. W.; Hanchanale, V.; Harji, D. P.; Helewa, R. M.; Heriot, A. G.; Hochman, D.; Hohenberger, W.; Holm, T.; Hompes, R.; Jenkins, J. T.; Kaffenberger, S.; Kandaswamy, G. V.; Kapur, S. |
| Publication Year: | 2020 |
| Collection: | Università degli Studi di Udine: CINECA IRIS |
| Subject Terms: | Artificial intelligence; artificial neural network; colorectal surgery; machine learning; pelvic exenteration |
| Description: | Aim: We aim to compare machine learning with neural network performance in predicting R0 resection (R0), length of stay > 14 days (LOS), major complication rates at 30 days postoperatively (COMP) and survival greater than 1 year (SURV) for patients having pelvic exenteration for locally advanced and recurrent rectal cancer. Method: A deep learning computer was built and the programming environment was established. The PelvEx Collaborative database was used which contains anonymized data on patients who underwent pelvic exenteration for locally advanced or locally recurrent colorectal cancer between 2004 and 2014. Logistic regression, a support vector machine and an artificial neural network (ANN) were trained. Twenty per cent of the data were used as a test set for calculating prediction accuracy for R0, LOS, COMP and SURV. Model performance was measured by plotting receiver operating characteristic (ROC) curves and calculating the area under the ROC curve (AUROC). Results: Machine learning models and ANNs were trained on 1147 cases. The AUROC for all outcome predictions ranged from 0.608 to 0.793 indicating modest to moderate predictive ability. The models performed best at predicting LOS > 14 days with an AUROC of 0.793 using preoperative and operative data. Visualized logistic regression model weights indicate a varying impact of variables on the outcome in question. Conclusion: This paper highlights the potential for predictive modelling of large international databases. Current data allow moderate predictive ability of both complex ANNs and more classic methods. |
| Document Type: | article in journal/newspaper |
| Language: | English |
| Relation: | info:eu-repo/semantics/altIdentifier/wos/WOS:000553038700001; volume:22; issue:12; firstpage:1933; lastpage:1940; numberofpages:8; journal:COLORECTAL DISEASE; https://hdl.handle.net/11390/1255292 |
| DOI: | 10.1111/codi.15235 |
| Availability: | https://hdl.handle.net/11390/1255292; https://doi.org/10.1111/codi.15235 |
| Accession Number: | edsbas.3DC6016A |
| Database: | BASE |