Decision tree (machine learning)
The decision tree model in machine learning is an algorithm that offers choices based on characteristics of the data. It follows 'branch node theory' in which each branch will represent a variable alongside a decision.
Often decision tree models will be expressed in the following rule format:
- IF variable = x AND variable = y THEN outcome = z
There are different types of decision tree model algorithms. Some of these algorithms have been used for radiology image classification and thus appear the radiology literature ( including CART 1 and CHAID 2 ) or radiology related computer science literature ( ID3 3,4). In contrast to many methods of machine learning, decision tree models utilize heuristics and are therefore comprehensible.
While classically visualized by diagrams of connected nodes, decision tree models can be visualized by partitioned graphs or perspective plots.
- 1: Fabio S Aguiar, Luciana L Almeida, Antonio Ruffino-Netto, Afranio Lineu Kritski, Fernanda CQ Mello, Guilherme L Werneck. Classification and regression tree (CART) model to predict pulmonary tuberculosis in hospitalized patients. (2012) BMC Pulmonary Medicine. 12 (1): 40. doi:10.1186/1471-2466-12-40 - Pubmed
- 2: Wengert GJ, Baltzer PAT, Bickel H, Thurner P, Breitenseher J, Lazar M, Pones M, Peck-Radosavljevic M, Hucke F, Ba-Ssalamah A. Differentiation of Intrahepatic Cholangiocellular Carcinoma from Hepatocellular Carcinoma in the Cirrhotic Liver Using Contrast-enhanced MR Imaging. (2017) Academic radiology. 24 (12): 1491-1500. doi:10.1016/j.acra.2017.06.005 - Pubmed
- 3: Angayarkanni, S.Pitchumani & Banu Kamal, Nadira. (2012). MRI mammogram image classification using ID3 algorithm. IET Conference Publications. 2012. 1-5. 10.1049/cp.2012.0464.
- 4: Wei CP, Hu PJ, Sheng OR. A knowledge-based system for patient image pre-fetching in heterogeneous database environments--modeling, design, and evaluation. (2001) IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society. 5 (1): 33-45. Pubmed
Related Radiopaedia articles
- artificial intelligence (AI)
- imaging data sets
- computer-aided diagnosis (CAD)
machine learning (overview)
- machine learning processes
- machine learning models
- common data preparation/preprocessing steps
- DICOM to bitmap conversion
- principal component analysis
- training, testing and validation datasets
- mean squared error
- cross entropy
- optimization algorithms
- stochastic gradient descent
- momentum (Nesterov)
- linear and quadratic
- batch normalization
- natural language processing
- rule-based expert systems