Machine learning (ML) is the subfield of artificial intelligence (AI), born from the marriage between statistics and computer science, with the unique purpose of building prediction algorithms able to improve their performances by automatically learning from massive data sets. The availability of ever-growing computational power and highly evolved pattern recognition software has led to the spread of ML-based systems able to perform complex tasks in bioinformatics, medical imaging, and diagnostics. These intelligent tools could be the answer to the unmet need for non-invasive and patient-tailored instruments for the diagnosis and management of bladder cancer (BC), which are still based on old technologies and unchanged nomograms. We reviewed the most significant evidence on ML in the diagnosis, prognosis, and management of bladder cancer, to find out if these intelligent technologies are ready to be introduced into the daily clinical practice of the urologist.
American Cancer Society. Cancer facts & figures 2020. Atlanta, GA: American Cancer Society, 2020, pp.10–21.
2.
RichtersAAbenKKHKiemeneyLALM. The global burden of urinary bladder cancer: an update. World J Urol2020; 38: 1895–1904.
3.
GoldenbergSLNirGSalcudeanSE.A new era: artificial intelligence and machine learning in prostate cancer. Nat Rev Urol2019; 16(7): 391–403.
4.
EminagaOEminagaNSemjonowA, et al. Diagnostic classification of cystoscopic images using deep convolutional neural networks. JCO Clin Cancer Inform2018; 2: 1–8.
5.
ShkolyarEJiaXChangTC, et al. Augmented bladder tumor detection using deep learning. Eur Urol2019; 76(6): 714–718.
6.
IkedaANosatoHKochiY, et al. Support system of cystoscopic diagnosis for bladder cancer based on artificial intelligence. J Endourol2020; 34(3): 352–358.
7.
LucasMLiemEIMLSavci-HeijinkCD, et al. Toward automated in vivo bladder tumor stratification using confocal laser endomicroscopy. J Endourol2019; 33(11): 930–937.
8.
PantazopoulosDKarakitsosPIokim-LiossiA, et al. Back propagation neural network in the discrimination of benign from malignant lower urinary tract lesions. J Urol1998; 159(5): 1619–1623.
9.
ParekattilSJFisherHAKoganBA.Neural network using combined urine nuclear matrix protein-22, monocyte chemoattractant protein-1 and urinary intercellular adhesion molecule-1 to detect bladder cancer. J Urol2003; 169(3): 917–920.
10.
ShaoCHChenCLLinJY, et al. Metabolite marker discovery for the detection of bladder cancer by comparative metabolomics. Oncotarget2017; 8(24): 38802–38810.
11.
SokolovIDokukinMEKalaparthiV, et al. Noninvasive diagnostic imaging using machine-learning analysis of nanoresolution images of cell surfaces: detection of bladder cancer. Proc Natl Acad Sci U S A2018; 115(51): 12920–12925.
12.
SpyridonosPCavourasDRavazoulaP, et al. Neural network-based segmentation and classification system for automated grading of histologic sections of bladder carcinoma. Anal Quant Cytol Histol2002; 24(6): 317–324.
13.
TasoulisDKSpyridonosPPavlidisNG, et al. Urinary bladder tumor grade diagnosis using on-line trained neural networks. In: VPaladeRJHowlettLJain (eds) Knowledge-based intelligent information and engineering systems. KES 2003. Lecture Notes in Computer Science, vol. 2773. Berlin, Heidelberg: Springer, 2003.
14.
WoerlACEcksteinMGeigerJ, et al. Deep learning predicts molecular subtype of muscle-invasive bladder cancer from conventional histopathological slides. Eur Urol2020; 78: 256–264.
15.
QureshiKNNaguibRNHamdyFC, et al. Neural network analysis of clinicopathological and molecular markers in bladder cancer. J Urol2000; 163(2): 630–633.
16.
FujikawaKMatsuiYKobayashiT, et al. Predicting disease outcome of non-invasive transitional cell carcinoma of the urinary bladder using an artificial neural network model: results of patient follow-up for 15 years or longer. Int J Urol2003; 10(3): 149–152.
17.
CattoJWLinkensDAAbbodMF, et al. Artificial intelligence in predicting bladder cancer outcome: a comparison of neuro-fuzzy modeling and artificial neural networks. Clin Cancer Res2003; 9(11): 4172–4177.
18.
CattoJWAbbodMFLinkensDA, et al. Neuro-fuzzy modeling: an accurate and interpretable method for predicting bladder cancer progression. J Urol2006; 175(2): 474–479.
19.
BassiPSaccoEDe MarcoV, et al. Prognostic accuracy of an artificial neural network in patients undergoing radical cystectomy for bladder cancer: a comparison with logistic regression analysis. BJU Int2007; 99(5): 1007–1012.
20.
BhambhvaniHPZamoraAShkolyarE, et al. Development of robust artificial neural networks for prediction of 5-year survival in bladder cancer. Urol Oncol. Epub ahead of print 24 June 2020. DOI:10.1016/j.urolonc.2020.05.009.
21.
CattoJWAbbodMFLinkensDA, et al. Neurofuzzy modeling to determine recurrence risk following radical cystectomy for nonmetastatic urothelial carcinoma of the bladder. Clin Cancer Res2009; 15(9): 3150–3155.
22.
BartschGJrMitraAPMitraSA, et al. Use of artificial intelligence and machine learning algorithms with gene expression profiling to predict recurrent nonmuscle invasive urothelial carcinoma of the bladder. J Urol2016; 195(2): 493–498.
23.
HasnainZMasonJGillK, et al. Machine learning models for predicting post-cystectomy recurrence and survival in bladder cancer patients. PLoS One2019; 14(2): e0210976.
24.
GarapatiSSHadjiiskiLChaKH, et al. Urinary bladder cancer staging in CT urography using machine learning. Med Phys2017; 44(11): 5814–5823.
25.
ZhangXXuXTianQ, et al. Radiomics assessment of bladder cancer grade using texture features from diffusion-weighted imaging. J Magn Reson Imaging2017; 46(5): 1281–1288.
26.
ChaKHHadjiiskiLMSamalaRK, et al. Bladder cancer segmentation in CT for treatment response assessment: application of deep-learning convolution neural network-a pilot study. Tomography2016; 2(4): 421–429.
27.
ChaKHHadjiiskiLChanHP, et al. Bladder cancer treatment response assessment in CT using radiomics with deep-learning. Sci Rep2017; 7(1): 8738.
28.
WuEHadjiiskiLMSamalaRK, et al. Deep learning approach for assessment of bladder cancer treatment response. Tomography2019; 5(1): 201–208.
29.
DeoRC.Machine learning in medicine. Circulation2015; 132(20): 1920–1930.
30.
AnagnostouTRemziMLykourinasM, et al. Artificial neural networks for decision-making in urologic oncology. Eur Urol2003; 43(6): 596–603.
31.
PodgorelecVKokolPStiglicB, et al. Decision trees: an overview and their use in medicine. J Med Syst2002; 26(5): 445–463.
ShenYJZhuYPYeDW, et al. Narrow-band imaging flexible cystoscopy in the detection of primary non-muscle invasive bladder cancer: a “second look” matters?Int Urol Nephrol2012; 44: 451–457.
34.
KauschISommerauerMMontorsiF, et al. Photodynamic diagnosis in non-muscle-invasive bladder cancer: a systematic review and cumulative analysis of prospective studies. Eur Urol2010; 57: 595–606.
35.
SchönebeckJ.Atlas of cystoscopy. New York, NY: Grune and Stratton, 1985.
36.
LotanYRoehrbornCG.Sensitivity and specificity of commonly available bladder tumor markers versus cytology: results of a comprehensive literature review and metaanalyses. Urology2003; 61(1): 109–118.
37.
VroomanOPWitjesJA.Urinary markers in bladder cancer. Eur Urol2008; 53: 909–916.
38.
OlsenLHOvergaardSFrederiksenP, et al. The reliability of staging and grading of bladder tumors. Impact of misinformation on the pathologist’s diagnosis. Scand J Urol Nephrol1993; 27(3): 349–353.
39.
RobertsonAGKimJAl-AhmadieH, et al. Comprehensive molecular characterization of muscle-invasive bladder cancer. Cell2017; 171: 540–556.e25.
40.
McConkeyDJChoiW.Molecular subtypes of bladder cancer. Curr Oncol Rep2018; 20: 1–7.
41.
SoukupVČapounOCohenD, et al. Risk stratification tools and prognostic models in non-muscle-invasive bladder cancer: a critical assessment from the European Association of Urology Non-muscle-invasive Bladder Cancer Guidelines Panel. Eur Urol Focus2020; 6(3): 479–489.
42.
WitjesJABruinsHMCathomasR, et al. European Association of Urology guidelines on muscle-invasive and metastatic bladder cancer: summary of the 2020 guidelines. Eur Urol2021; 79(1): 82–104.
43.
SylvesterRJvan der MeijdenAPOosterlinckW, et al. Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials. Eur Urol2006; 49(3): 466–477.
44.
BurkeHBGoodmanPHRosenDB, et al. Artificial neural networks improve the accuracy of cancer survival prediction. Cancer1997; 79(4): 857–862.
45.
UenoHMurphyJJassJR, et al. Tumor ‘budding’ as an index to estimate the potential of aggressiveness in rectal cancer. Histopathology2002; 40(2): 127–132.
46.
FukumotoKKikuchiEMikamiS, et al. Tumor budding, a novel prognostic indicator for predicting stage progression in T1 bladder cancers. Cancer Sci2016; 107(9): 1338–1344.
47.
BrieuNGavrielCGNearchouIP, et al. Automated tumor budding quantification by machine learning augments TNM staging in muscle-invasive bladder cancer prognosis. Sci Rep2019; 9(1): 5174.
48.
LambinPRios-VelazquezELeijenaarR, et al. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer2012; 48(4): 441–446.
49.
ParmarCRios VelazquezELeijenaarR, et al. Robust Radiomics feature quantification using semiautomatic volumetric segmentation. PLoS One2014; 9(7): e102107.
50.
DuanCLiangZBaoS, et al. A coupled level set framework for bladder wall segmentation with application to MR cystography. IEEE Trans Med Imaging2010; 29: 903–915.
51.
DolzJXuXRonyJ, et al. Multiregion segmentation of bladder cancer structures in MRI with progressive dilated convolutional networks. Med Phys2018; 45(12): 5482–5493.
