Classification of Soft Tissue Tumors by Machine Learning Algorithms

Juntu, Jaber; Schepper, A. M. De; Dyck, Pieter Van; Dyck, D. Van; Gielen, Jan; Parizel, Paul M.; Sijbers, Jan

doi:10.5772/27757

Cited by 5 publications

(11 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, complex MRI data for tumors from clinical analyses or synthetic databases are difficult for physicians to interpret [2] . The MRI devices and protocols used for imaging pricing can vary from one scan to another, imposing intensity bias and other variations on each different image slice in the dataset.…”

Section: Challenge Of ML For Soft Tissues Tumor Classificationmentioning

confidence: 99%

“…In general, resampling is necessary when the dataset has been recorded in a given time interval or number of samples and we want to modify this interval or number of samples [2] . The impact of this strategy is also visible on the data distribution, making it more suitable for the classification algorithms.…”

Section: Resampling Techniquesmentioning

confidence: 99%

“…In general, resampling is necessary when the dataset has been recorded in a given time interval or includes a number of samples and we want to modify this interval or number of samples [2] . In this case, we want to resize and reorder the data in order to optimize the model's performances by eliminating the bias and the variance of the algorithms to build a stable and realistic model.…”

Section: Resampling the Data To Improve Performancementioning

confidence: 99%

“…As the name suggests, these are sensitive tissues that can be affected by several infections, including tumors that can develop almost anywhere in the human body. The malignant types of these tumors, also known as Soft Tissue Sarcomas (STS), are grouped together because they share many microscopic features, exhibit the same symptoms, and are almost similarly treated [1,2] . Yet, effective diagnosis of Soft Tissues Tumors (STT) is still a big challenge owing to the difficulty in detecting these cancers.…”

Section: Introductionmentioning

confidence: 99%

“…MRI is currently considered the standard diagnostic tool for the detection and classification of STT [3] with well characterized biological properties such as cellular origins and tumour specimens [4] used to distinguish tumors. MRI can be used to analyze textural characteristics or other less characterized tumor characteristics (average MRI signal intensity, shape of tumor boundaries) for several reasons: (1) ease of computation textural characteristics, (2) wide correlation of textural characteristics to tumor pathology [5,6] , and (3) robustness to changes in MRI acquisition parameters such as changes in the resolution of the tumor image and the corruption of the MRI image due to heterogeneity of the magnetic field [2] . Such magnetic field heterogeneity in MRI makes it difficult to perceive the texture in certain malignant tumors and humans have a limited capacity to perceive and discriminate these textures as well [7] .…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Improvement in automated diagnosis of soft tissues tumors using machine learning

Alaoui¹,

Tékouabou

Hartini

et al. 2021

Big Data Min. Anal.

View full text Add to dashboard Cite

Soft Tissue Tumors (STT) are a form of sarcoma found in tissues that connect, support, and surround body structures. Because of their shallow frequency in the body and their great diversity, they appear to be heterogeneous when observed through Magnetic Resonance Imaging (MRI). They are easily confused with other diseases such as fibroadenoma mammae, lymphadenopathy, and struma nodosa, and these diagnostic errors have a considerable detrimental effect on the medical treatment process of patients. Researchers have proposed several machine learning models to classify tumors, but none have adequately addressed this misdiagnosis problem. Also, similar studies that have proposed models for evaluation of such tumors mostly do not consider the heterogeneity and the size of the data. Therefore, we propose a machine learning-based approach which combines a new technique of preprocessing the data for features transformation, resampling techniques to eliminate the bias and the deviation of instability and performing classifier tests based on the Support Vector Machine (SVM) and Decision Tree (DT) algorithms. The tests carried out on dataset collected in Nur Hidayah Hospital of Yogyakarta in Indonesia show a great improvement compared to previous studies. These results confirm that machine learning methods could provide efficient and effective tools to reinforce the automatic decision-making processes of STT diagnostics.

show abstract

Section: Challenge Of ML For Soft Tissues Tumor Classificationmentioning

confidence: 99%

Section: Resampling Techniquesmentioning

confidence: 99%

Section: Resampling the Data To Improve Performancementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%