Machine learning models predict the primary sites of head and neck squamous cell carcinoma metastases based on <scp>DNA</scp> methylation

Leitheiser, Maximilian; Capper, David; Seegerer, Philipp; Lehmann, Annika; Schüller, Ulrich; Müller, Klaus‐Robert; Klauschen, Frederick; Jurmeister, Philipp; Bockmayr, Michael

doi:10.1002/path.5845

Cited by 32 publications

(23 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All the foods we eat are genetically modified. From chemical additives, multi-colored harmful plastic containers in the name of modular kitchen, food prepared in the oven, Teflon coating on non-stick dosa to drinking milk, there is a risk of cancer in everything [2]. Beauty face paints, hair dyes, powders, sunscreen lotions...…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

deep DNA machine learning model to classify the tumor genome of patients with tumor sequencing

Logeshwaran¹,

Adhikari

Joshi³

et al. 2022

ijhs

156

View full text Add to dashboard Cite

In general, the various medical systems currently available provide insights into changes in the tumor genome of patients with tumor sequencing. Most of the tumor DNA sequencing can also be referred to as genetic specification or genetic testing. The sequence results help clinical decision-making to develop a personalized cancer treatment plan based on the molecular characteristics of the tumor rather than a one-size-fits-all treatment approach. The tumor sequencing also plays a major role in cancer research. In this paper, an improved method based on machine learning was proposed to analyze the sequencing and tumor sequencing patterns of the human gene. This proposed method analyzes the circulatory problems of patients with different tumor types for analysis in the public domain. It also constantly monitors large data sets of cancer or tumor genetic sequences to calculate tumor size and location. This allows the doctor to get an accurate report on the type of tumor and the problems it can cause to the patient. The Analysis of these datasets of cancer tumor gene sequences reveals that the genetic makeup of each patient is different and that no two cancers are the same.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

“…Cancer treatment doctors say that early stage disease can be completely cured. In the changing chemical world, new diseases are emerging [2]. At the same time, there is no escape from it without awareness, basic clarity and self-defense.…”

Section: Introductionmentioning

confidence: 99%

deep DNA machine learning model to classify the tumor genome of patients with tumor sequencing

Logeshwaran¹,

Adhikari

Joshi³

et al. 2022

ijhs

156

View full text Add to dashboard Cite

show abstract

“…Most publications on similar classi cation tasks reported the successful use of a single RF model with additional calibration methods [35,36]. However, in a comparison of the LOGREG, NN, SVM, and RF models for predicting the primary site of HNSC tumors from their DNA methylation pro le, the RF model performed considerably worse than the other three classi ers [21]. A comparison of a multitude of methods for the characterization and prediction of HPV in patients with OPCs showed that a generalized linear model (with a logistic kernel) performed best among all methods and was substantially better than the RF model [37].…”

Section: Discussionmentioning

confidence: 99%

“…Deep learning techniques applied to medical image analysis can integrate qualitative and quantitative imaging features to construct an optimal diagnostic algorithm prediction model with high diagnostic accuracy [20]. Models based on DNA methylation pro les [21] and cancer genome information [22] have been used to predict the origin of cancers of unknown primary. However, experience in the application of ML techniques based on radiomic features from imaging of head and neck cancer is limited.…”

Section: Introductionmentioning

confidence: 99%

Title: Discriminating between p16-negative oropharyngeal and non-oropharyngeal origins by their metastatic lymph nodes using machine learning approach based on MRI radiomics

Bao

Zhao

et al. 2022

Preprint

View full text Add to dashboard Cite

Objectives:To examine the capability of MRI radiomic analysis to differentiate oropharyngeal and non-oropharyngeal origins of cervical metastatic lymph nodes (LNs) in p16-negative patients following machine-learning approaches. Methods:A total of 150 patients (training set: 90, validation set: 60) diagnosed with p16-negative head and neck squamous cell carcinoma with cervical metastatic LNs were retrospectively included. Radiomic features were extracted from pretreatment diagnostic T2-weighted fat-suppressed and contrast-enhanced T1-weighted images of the cervical metastatic LNs. The Wilcoxon-Mann-Whitney test (with p < 0.05) and absolute biserial correlation (> 0.48) were used to identify the discriminant features between the oropharyngeal and non-oropharyngeal groups. The discriminant features were used to train and test four different machine-learning classifiers. The discriminative performances of classifiers were evaluated by AUC, sensitivity, specificity, PPV, and NPV and compared. Results:Five radiomic features were significantly different (P < 0.05, |biserial correlation| > 0.48) between the p16-negative oropharyngeal and non-oropharyngeal groups. Among the four classifiers trained and applied for discrimination between oropharyngeal and non-oropharyngeal origins, the logistic regression and neural network both showed the highest prediction power, with AUC of 0.87 and 0.74 in the training and validation cohorts, respectively. The accuracy, sensitivity, specificity, PPV, and NPV of the two classifiers in the validation cohort were 0.72, 0.82, 0.66, 0.58, and 0.86, respectively. Conclusion:Radiomics and machine-learning diagnostic models with MRI data are potentially useful supportive tools to discriminate between oropharyngeal and non-oropharyngeal origins in p16-negative patients with cervical metastatic LNs.

show abstract

“…In recent years, machine learning and data mining have contributed to the biomarker screening for early detection, diagnosis, drug application, metastases, and prognosis of various cancers (9)(10)(11). For example, by employing four machine learning models, Leitheiser et al (12) predicted the primary sites of HNSC metastases based on DNA methylation. Rendleman et al (13) developed a machine learning method to improve the usability of the HNSC dataset for enhancing future oncological decisions and clinical applications.…”

Section: Introductionmentioning

confidence: 99%

Data mining on identifying diagnosis and prognosis biomarkers in head and neck squamous carcinoma

Yao

Zhao

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: Head and neck squamous carcinoma (HNSC) is cancer with a poor prognosis that induces high cancer-related death worldwide. The biomarker screening on diagnosis and prognosis is of great importance. This research is aimed to explore the specific diagnostic and prognostic biomarkers for HNSC through bioinformatics analysis.Results: The mutation and dysregulation data were acquired from UCSC Xena and TCGA databases. The top ten genes with mutation frequency in HNSC were TP53 (66%), TTN (35%), FAT1 (21%), CDKN2A (20%), MUC16 (17%), CSMD3 (16%), PIK3CA (16%), NOTCH1 (16%), SYNE1 (15%), LRP1B (14%). A total of 1,060 DEGs were identified, with 396 up-regulated and 665 downregulated in HNSC patients. Patients with lower expression of ACTN2 (P=0.039, HR=1.3), MYH1 (P=0.005, HR=1.5), MYH2 (P=0.035, HR=1.3), MYH7 (P=0.053, HR=1.3), and NEB (P=0.0043, HR=1.5) exhibit longer overall survival time in HNSC patients. The main DEGs were further analyzed by pan-cancer expression and immune cell infiltration analyses. We can observe that MYH1, MYH2, and MYH7 were dysregulated in the cancers. Compared with HNSC, their expression levels are lower in the other types of cancers. All the five DEGs have a significant positive correlation with CD4+ T cells and macrophages.Conclusion: ACTN2, MYH1, MYH2, MYH7, and NEB genes were significantly associated with poor prognosis in HNSC. Moreover, MYH1, MYH2, and MYH7 were expected to be the specific diagnostic and prognostic molecular biomarkers of HNSC.

show abstract

Machine learning models predict the primary sites of head and neck squamous cell carcinoma metastases based on DNA methylation

Cited by 32 publications

References 41 publications

deep DNA machine learning model to classify the tumor genome of patients with tumor sequencing

deep DNA machine learning model to classify the tumor genome of patients with tumor sequencing

Title: Discriminating between p16-negative oropharyngeal and non-oropharyngeal origins by their metastatic lymph nodes using machine learning approach based on MRI radiomics

Data mining on identifying diagnosis and prognosis biomarkers in head and neck squamous carcinoma

Contact Info

Product

Resources

About