User Tools

Site Tools


machine_learning

Machine learning

Machine learning (ML) is a domain of artificial intelligence that allows computer algorithms to learn patterns by studying data directly without being explicitly programmed 1) 2).

ML methods are already widely applied in multiple aspects of our daily lives, although this is not always obvious to the casual observer; common examples are email spam filters, search suggestions, online shopping suggestions, and speech recognition in smartphones 3).


ML is increasingly tested in neurosurgical applications and even demonstrated to emulate the performance of clinical experts 4) 5) 6) 7) 8) 9) 10) 11) 12) 13) 14) 15) 16) 17) 18) 19) 20) 21) 22) 23) 24) 25) 26).


Automated analysis of radiological data for diagnosis, segmentation, or outcome prediction could, be one of the first ML applications that finds its way to actual clinical practice 27).


A systematic search was performed in the PubMed and Embase databases as of August 2016 to review all studies comparing the performance of various ML approaches with that of clinical experts in neurosurgical literature.

Twenty-three studies were identified that used ML algorithms for diagnosis, presurgical planning, or outcome prediction in neurosurgical patients. Compared to clinical experts, ML models demonstrated a median absolute improvement in accuracy and area under the receiver operating curve of 13% (interquartile range 4-21%) and 0.14 (interquartile range 0.07-0.21), respectively. In 29 (58%) of the 50 outcome measures for which a P-value was provided or calculated, ML models outperformed clinical experts (P < .05). In 18 of 50 (36%), no difference was seen between ML and expert performance (P > .05), while in 3 of 50 (6%) clinical experts outperformed ML models (P < .05). All 4 studies that compared clinicians assisted by ML models vs clinicians alone demonstrated a better performance in the first group.

Senders et al., conclude that ML models have the potential to augment the decision-making capacity of clinicians in neurosurgical applications; however, significant hurdles remain associated with creating, validating, and deploying ML models in the clinical setting. Shifting from the preconceptions of a human-vs-machine to a human-and-machine paradigm could be essential to overcome these hurdles 28).


Lazaridis et al., and others, have developed predictive models based on machine learning from continuous time series of intracranial pressure and partial pressure of brain tissue oxygen. These models provide accurate predictions of physiologic crises events in a timely fashion, offering the opportunity for an earlier application of targeted interventions.They review the rationale for prediction, discuss available predictive models with examples, and offer suggestions for their future prospective testing in conjunction with preventive clinical algorithms 29).


Machine learning (ML) is a domain of artificial intelligence that allows computer algorithms to learn from experience without being explicitly programmed.

To summarize neurosurgical applications of ML where it has been compared to clinical expertise, here referred to as “natural intelligence.”

Two important and rapidly developing scientific movements—data reproducibility and machine learning—are central to a recent Neuron paper by Chung et al 30)


A systematic search was performed in the PubMed and Embase databases as of August 2016 to review all studies comparing the performance of various ML approaches with that of clinical experts in neurosurgical literature.

Twenty-three studies were identified that used ML algorithms for diagnosis, presurgical planning, or outcome prediction in neurosurgical patients. Compared to clinical experts, ML models demonstrated a median absolute improvement in accuracy and area under the receiver operating curve of 13% (interquartile range 4-21%) and 0.14 (interquartile range 0.07-0.21), respectively. In 29 (58%) of the 50 outcome measures for which a P -value was provided or calculated, ML models outperformed clinical experts ( P < .05). In 18 of 50 (36%), no difference was seen between ML and expert performance ( P > .05), while in 3 of 50 (6%) clinical experts outperformed ML models ( P < .05). All 4 studies that compared clinicians assisted by ML models vs clinicians alone demonstrated a better performance in the first group.

We conclude that ML models have the potential to augment the decision-making capacity of clinicians in neurosurgical applications; however, significant hurdles remain associated with creating, validating, and deploying ML models in the clinical setting. Shifting from the preconceptions of a human-vs-machine to a human-and-machine paradigm could be essential to overcome these hurdles 31).


Yepes-Calderon et al. presented a segmentation strategy based on an algorithm that uses four features extracted from the medical images to create a statistical estimator capable of determining ventricular volume. When compared with manual segmentations, the correlation was 94% and holds promise for even better accuracy by incorporating the unlimited data available. The volume of any segmentable structure can be accurately determined utilizing the machine learning strategy presented and runs fully automatically within the PACS 32).

1)
Mitchell, TM. Machine Learning . Vol. 1. New York: McGraw-Hill Science/Engineering/Math; 1997.
2)
Noble WS. What is a support vector machine? Nat Biotechnol . 2006;24(12):1565-1567.
3)
Jordan MI, Mitchell TM. Machine learning: trends, perspectives, and prospects. Science . 2015;349(6245):255-260.
4)
Mariak Z, Swiercz M, Krejza J, Lewko J, Lyson T. Intracranial pressure processing with artificial neural networks: classification of signal properties. Acta Neurochir (Wien) . 2000;142(4):407-411; discussion 411-402.
5)
Nucci CG, De Bonis P, Mangiola A et al. Intracranial pressure wave morphological classification: automated analysis and clinical validation. Acta Neurochir (Wien) . 2016;158(3):581-588; discussion 588.
6)
Sieben G, Praet M, Roels H, Otte G, Boullart L, Calliauw L. The development of a decision support system for the pathological diagnosis of human cerebral tumours based on a neural network classifier. Acta Neurochir (Wien) . 1994;129(3-4):193-197.
7)
Mathew B, Norris D, Mackintosh I, Waddell G. Artificial intelligence in the prediction of operative findings in low back surgery. Brit J Neurosurg . 1989;3(2):161-170.
8)
Arle JE, Perrine K, Devinsky O, Doyle WK. Neural network analysis of preoperative variables and outcome in epilepsy surgery. J Neurosurg . 1999;90(6):998-1004.
9)
Gazit T, Andelman F, Glikmann-Johnston Y et al. Probabilistic machine learning for the evaluation of presurgical language dominance. J Neurosurg . 2016;125(2):1-13.
10)
Shi HY, Hwang SL, Lee KT, Lin CL. In-hospital mortality after traumatic brain injury surgery: a nationwide population-based comparison of mortality predictors used in artificial neural network and logistic regression models. J Neurosurg . 2013;118(4):746-752.
11)
Azimi P, Mohammadi HR. Predicting endoscopic third ventriculostomy success in childhood hydrocephalus: an artificial neural network analysis. J Neurosurg Pediatr . 2014;13(4):426-432.
12)
Azimi P, Mohammadi H. Prediction of successful ETV outcome in childhood hydrocephalus: an artificial neural networks analysis. J Neurosurg . 2015;122(6):426-432.
13)
Chang K, Zhang B, Guo X et al. Multimodal imaging patterns predict survival in recurrent glioblastoma patients treated with bevacizumab. Neuro-oncology . 2016;18(12):1680-1687.
14)
Jones TL, Byrnes TJ, Yang G, Howe FA, Bell BA, Barrick TR. Brain tumor classification using the diffusion tensor image segmentation (D-SEG) technique. Neuro-oncology . 2015;17(3):466-476.
15)
Macyszyn L, Akbari H, Pisapia JM et al. Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques. Neuro-oncology . 2016;18(3):417-425.
16)
Teplyuk NM, Mollenhauer B, Gabriely G et al. MicroRNAs in cerebrospinal fluid identify glioblastoma and metastatic brain cancers and reflect disease activity. Neuro-oncology . 2012;14(6):689-700.
17)
Zhang B, Chang K, Ramkissoon S et al. Multimodal MRI features predict isocitrate dehydrogenase genotype in high-grade gliomas. Neuro-oncology . 2017;19(1):109-117.
18)
Fouke SJ, Weinberger K, Kelsey M et al. A machine-learning-based classifier for predicting a multi-parametric probability map of active tumor extent within glioblastoma multiforme. Neuro-oncology . 2012;14:vi124-vi125.
19)
Kim LM, Commean P, Boyd A et al. Predicting the location and probability of viable tumor within glioblastoma multiforme with multiparametric magnetic resonance imaging. Neuro-oncology . 2012;14:vi120-vi128.
20)
Orphanidou-Vlachou E, Vlachos N, Davies N, Arvanitis T, Grundy R, Peet A. Texture analysis of T1-and t2-weighted magnetic resonance images to discriminate posterior fossa tumors in children. Neuro-oncology . 2014;16:i123-i126.
21)
Rayfield C, Swanson K. Predicting the response to treatment in GBM: Machine learning on clinical images. Neuro-oncology . 2015;17:v167.
22)
Akbari H, Macyszyn L, Da X et al. Imaging surrogates of infiltration obtained via multiparametric imaging pattern analysis predict subsequent location of recurrence of glioblastoma. Neurosurgery . 2016;78(4):572-580.
23)
Mitchell TJ, Hacker CD, Breshears JD et al. A novel data-driven approach to preoperative mapping of functional cortex using resting-state functional magnetic resonance imaging. Neurosurgery . 2013;73(6):969-982; discussion 982-963.
24)
Oermann EK, Kress MA, Collins BT et al. Predicting survival in patients with brain metastases treated with radiosurgery using artificial neural networks. Neurosurgery . 2013;72(6):944-951; discussion 952.
25)
Taghva A. An automated navigation system for deep brain stimulator placement using hidden Markov models. Neurosurgery . 2010;66(3 Suppl Operative):108-117; discussion 117.
26)
Dumont TM, Rughani AI, Tranmer BI. Prediction of symptomatic cerebral vasospasm after aneurysmal subarachnoid hemorrhage with an artificial neural network: feasibility and comparison with logistic regression models. World Neurosurg . 2011;75(1):57-63; discussion 25-58.
27)
Obermeyer Z, Emanuel EJ. Predicting the future - big data, machine learning, and clinical medicine. N Engl J Med . 2016;375(13):1216-1219.
28)
Senders JT, Arnaout O, Karhade AV, Dasenbrock HH, Gormley WB, Broekman ML, Smith TR. Natural and Artificial Intelligence in Neurosurgery: A Systematic Review. Neurosurgery. 2018 Aug 1;83(2):181-192. doi: 10.1093/neuros/nyx384. PubMed PMID: 28945910.
29)
Lazaridis C, Rusin CG, Robertson CS. Secondary Brain Injury: Predicting and Preventing Insults. Neuropharmacology. 2018 Jun 6. pii: S0028-3908(18)30279-X. doi: 10.1016/j.neuropharm.2018.06.005. [Epub ahead of print] Review. PubMed PMID: 29885419.
30)
Chung JE, Magland JF, Barnett AH, Tolosa VM, Tooker AC, Lee KY, Shah KG, Felix SH, Frank LM, Greengard LF. A Fully Automated Approach to Spike Sorting. Neuron. 2017 Sep 13;95(6):1381-1394.e6. doi: 10.1016/j.neuron.2017.08.030. PubMed PMID: 28910621; PubMed Central PMCID: PMC5743236.
31)
Senders JT, Arnaout O, Karhade AV, Dasenbrock HH, Gormley WB, Broekman ML, Smith TR. Natural and Artificial Intelligence in Neurosurgery: A Systematic Review. Neurosurgery. 2017 Sep 7. doi: 10.1093/neuros/nyx384. [Epub ahead of print] PubMed PMID: 28945910.
32)
Yepes-Calderon F, Nelson MD, McComb JG. Automatically measuring brain ventricular volume within PACS using artificial intelligence. PLoS One. 2018 Mar 15;13(3):e0193152. doi: 10.1371/journal.pone.0193152. eCollection 2018. PubMed PMID: 29543817.
machine_learning.txt · Last modified: 2018/07/21 13:06 by administrador