Machine learning to predict occult nodal metastasis in early oral squamous cell carcinoma

Bur, Andrés M.; Holcomb, Andrew J.; Goodwin, Sara; Woodroof, Janet; Karadaghy, Omar A.; Shnayder, Yelizaveta; Kakarala, Kiran; Brant, Jason A.; Shew, Matthew

doi:10.1016/j.oraloncology.2019.03.011

Cited by 115 publications

(98 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the available studies, the only one adopting these cutoff values (Faisal et al, ) reported odds ratio of 1.69 (≤5 mm vs. >5 mm and ≤10 mm), 2.15 (>5 mm and ≤10 mm vs. >10 mm), and 3.63 (≤5 mm vs. >10 mm) for occult lymph node metastasis. Interestingly, one recent study opens a new avenue toward the usefulness of machine learning algorithms in the prediction of occult lymph node metastasis in early‐stage OSCC (Bur et al, ). In this study, the machine learning algorithms outperformed DOI in predicting occult lymph node metastasis, with higher sensitivity and specificity (Bur et al, ).…”

Section: Discussionmentioning

confidence: 99%

“…Interestingly, one recent study opens a new avenue toward the usefulness of machine learning algorithms in the prediction of occult lymph node metastasis in early‐stage OSCC (Bur et al, ). In this study, the machine learning algorithms outperformed DOI in predicting occult lymph node metastasis, with higher sensitivity and specificity (Bur et al, ).…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Tumor depth of invasion and prognosis of early‐stage oral squamous cell carcinoma: A meta‐analysis

et al. 2019

View full text Add to dashboard Cite

Objectives To assess the prognosis for early‐stage oral squamous cell carcinoma according to tumor depth of invasion (DOI). Methods This study was logged in the PROSPERO database under protocol # CRD42017059976. The search was conducted in six electronic databases up to May 2019. Fixed‐effects meta‐analysis was performed for the calculation of the odds ratio (OR) and respective 95% CI. Primary outcomes were lymph node metastasis, recurrence, and survival. Heterogeneity was calculated by the I2 test. The certainty of evidence was assessed by Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach. Results Twenty‐seven studies were included (19 in the meta‐analysis) with 2,404 patients with a mean of 60 years of age. High tumor DOI is associated with a greater chance of presenting lymph node metastasis, regardless of the cutoff point for DOI (13 meta‐analysis; OR 1.69–53.08), recurrence (five meta‐analysis; OR 1.22–3.83), and lower chance of survival (1 meta‐analysis; OR 0.49). The certainty of evidence varied from very low to low. Conclusions Tumor DOI is a good prognosticator for early‐stage OSCC. The findings of the current meta‐analysis highlight the clinical relevance of DOI and corroborate its incorporation for staging OSCC.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Tumor depth of invasion and prognosis of early‐stage oral squamous cell carcinoma: A meta‐analysis

et al. 2019

View full text Add to dashboard Cite

show abstract

“…8 Machine learning approaches applying different datasets have recently been proposed to improve the classification of primary cancers and metastasis samples, as well as to predict cancer metastasis. 9, 10 Tapak L et al have found that the random forest (RF) has the highest specificity, the Naive Bayes (NB) has highest sensitivity while the traditional machine learning approaches [logistic regression (LR) and linear discriminant analysis] had the highest total accuracy for metastasis prediction in breast cancer. 11 In addition, the support vector machine (SVM) outperformed other machine learning methods for breast cancer survival prediction.…”

Section: Introductionmentioning

confidence: 99%

Predicting biomarkers from classifier for liver metastasis of colorectal adenocarcinomas using machine learning models

Han

Zhou

et al. 2020

Cancer Medicine

View full text Add to dashboard Cite

Background Early diagnosis of liver metastasis is of great importance for enhancing the survival of colorectal adenocarcinoma (CAD) patients, and the combined use of a single biomarker in a classier model has shown great improvement in predicting the metastasis of several types of cancers. However, it is little reported for CAD. This study therefore aimed to screen an optimal classier model of CAD with liver metastasis and explore the metastatic mechanisms of genes when applying this classier model. Methods The differentially expressed genes between primary CAD samples and CAD with metastasis samples were screened from the Moffitt Cancer Center (MCC) dataset http://GSE131418. The classification performances of six selected algorithms, namely, LR, RF, SVM, GBDT, NN, and CatBoost, for classification of CAD with liver metastasis samples were compared using the MCC dataset http://GSE131418 by detecting their classification test accuracy. In addition, the consortium datasets of http://GSE131418 and http://GSE81558 were used as internal and external validation sets to screen the optimal method. Subsequently, functional analyses and a drug‐targeted network construction of the feature genes when applying the optimal method were conducted. Results The optimal CatBoost model with the highest accuracy of 99%, and an area under the curve of 1, was screened, which consisted of 33 feature genes. A functional analysis showed that the feature genes were closely associated with a “steroid metabolic process” and “lipoprotein particle receptor binding” (eg APOB and APOC3). In addition, the feature genes were significantly enriched in the “complement and coagulation cascade” pathways (eg FGA, F2, and F9). In a drug‐target interaction network, F2 and F9 were predicted as targets of menadione. Conclusion The CatBoost model constructed using 33 feature genes showed the optimal classification performance for identifying CAD with liver metastasis.

show abstract

“…Depth of invasion of tumor is the most important variant to predict occult metastasis. Depth of invasion between 2 to 5 mm is an indication for elective node dissection [14]. When there is a nodal involvement in oral cancer neck dissection along with tumor removal is indicated, Lymph node excisional biopsies have been performed before neck dissection in primary oral cancer cases.…”

Section: Introductionmentioning

confidence: 99%

Association of Clinicopathological Features with Lymph Node Metastasis: A Cross Sectional Study of Oral Squamous Cell Carcinoma Patients

Abidi

Hosein

Butt

et al. 2020

JAMMR

View full text Add to dashboard Cite

Aims:The aim of this study is to determine association between lymph node metastasis and various clinico-pathological parameters in order to predict their future prognosis of cancer patients. Study Design: Cross sectional study design. Place and Duration of Study: Ziauddin Hospital, North Nazimabad campus, Karachi. Methodology: November 2018 to September 2019. Results: The present study enrolled 140 OSCC patients visiting Ziauddin University of which 30 (21.4%) were females and 110 (78.6%) were male patients. The mean age of patients was 46.99 ± 12.26. The buccal mucosa was the most common site observed 93 (66.4%). The mean size of tumour was 4.05 ± 2.261 mm, mean thickness was 1.70 ±1.32 mm. It was found that lymph node metastasis was not present in 66 (47.1%) of OSCC cases, 22 (15.7%) cases were found N1, 9(6.4%) N2a, 27(19.3%) N2b, 3 (2.1%) and 13(9.3%) N3. Abidi et al.; JAMMR, 32(2): 46-53, 2020; Article no.JAMMR.54793 47 Significant association was found between age and lymph node metastasis: (P 0.02); however, gender showed no significant association. (P 0.149). Thickness of tumor, site of tumor, peri-neural invasion, lympho-vascular invasion and bone penetration showed significant association with lymph node metastasis with (P 0.040, 0.021, 0.016, 0.00 and 0.001) respectively. Conclusion: Our study showed significant association of lymph node involvement with tumour staging, tumour thickness, lymphovascular involvement, perineural invasion and bone penetration. Our suggestion is to incorporate these clinic-pathological variables while determining the treatment and prognosis of Oral squamous cell Carcinoma (OSCC). Given the very high incidence and prevalence of Oral Cancer in our country and the region further multi-centered researches should be carried out and with larger sample sizes. Early diagnosis and restrictions on smokeless tobacco products can limit the occurrence and progression of OSCC. Original Research Article

show abstract

Machine learning to predict occult nodal metastasis in early oral squamous cell carcinoma

Cited by 115 publications

References 31 publications

Tumor depth of invasion and prognosis of early‐stage oral squamous cell carcinoma: A meta‐analysis

Tumor depth of invasion and prognosis of early‐stage oral squamous cell carcinoma: A meta‐analysis

Predicting biomarkers from classifier for liver metastasis of colorectal adenocarcinomas using machine learning models

Association of Clinicopathological Features with Lymph Node Metastasis: A Cross Sectional Study of Oral Squamous Cell Carcinoma Patients

Contact Info

Product

Resources

About