Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Non-linear effects in dose-time analyses: application of modern statistical techniques

J M Taylor1, F Zhang, H R Withers

  • 1Department of Radiation Oncology, UCLA School of Medicine, Los Angeles, CA 90095-1772, USA.

Radiotherapy and Oncology : Journal of the European Society for Therapeutic Radiology and Oncology
|January 10, 1998
PubMed
Summary

Generalized additive models reveal linear dose effects but suggest non-linear treatment time effects in head and neck radiotherapy outcomes. These statistical methods offer flexible exploration of complex relationships in experimental data.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

[Cardiac syndrome X and insulin resistance].

Zhonghua nei ke za zhi·2002
Same author

[The clinical significance of changes in plasma levels of alpha-granule membrane protein and D-dimer in patients with unstable angina].

Zhonghua nei ke za zhi·2002
Same author

[Value of P-glycoprotein and glutathione S-transferase-pi as chemo-resistant indicators in ovarian cancers].

Zhonghua zhong liu za zhi [Chinese journal of oncology]·2002
Same author

[Current status of pain relief with Chinese herbal drugs in patients with cancer].

Zhongguo Zhong xi yi jie he za zhi Zhongguo Zhongxiyi jiehe zazhi = Chinese journal of integrated traditional and Western medicine·2002
Same author

[Investigation of urinary interleukin-6 level in chronic renal failure patients and the influence of Rheum palmatum in treating it].

Zhongguo Zhong xi yi jie he za zhi Zhongguo Zhongxiyi jiehe zazhi = Chinese journal of integrated traditional and Western medicine·2002
Same author

Accelerating flap maturation by vascular endothelium growth factorin a rat tube flap model.

British journal of plastic surgery·2002

Area of Science:

  • Radiation Oncology
  • Biostatistics
  • Medical Physics

Background:

  • Investigating non-linear relationships between radiotherapy parameters and patient outcomes is crucial for optimizing treatment.
  • Fractionated radiotherapy for head and neck cancers involves complex interactions between dose, time, and biological response.

Purpose of the Study:

  • To illustrate the application of recently developed statistical techniques, specifically generalized additive models, for analyzing non-linear effects in radiotherapy.
  • To explore the non-linear effects of overall treatment time and total dose on the outcomes of fractionated head and neck radiotherapy.

Main Methods:

  • Application of generalized additive models to data from the patterns of fractionation study of tonsil cancer.
  • Utilizing data-driven methods to model non-linear relationships between predictor variables (dose, time, age) and outcome variables (local recurrence, latency time).

Related Experiment Videos

  • Replacing standard linear model components with smooth, non-linear functions estimated from the data (e.g., S1(dose) + S2(time)).

    • Main Results:

    • The effect of total dose on local recurrence probability was found to be linear.
    • A non-linear effect of overall treatment time on local recurrence was suggested.
    • No significant effect of dose and time on latency to recurrence was observed, but patient age showed a weak non-linear association with earlier recurrence in younger patients.

    Conclusions:

    • Generalized additive models offer a powerful and flexible approach for uncovering non-linear patterns in complex experimental data.
    • These statistical methods enhance the accuracy of regression models by allowing predictor variables to have non-linear effects on outcomes.