Statistics for Radiology Research

Abstract

Biostatistics is an essential component in most original research studies in imaging. In this article we discuss five key statistical concepts for study design and analyses in modern imaging research: statistical hypothesis testing, particularly focusing on noninferiority studies; imaging outcomes especially when there is no reference standard; dealing with the multiplicity problem without spending all your study power; relevance of confidence intervals in reporting and interpreting study results; and finally tools for assessing quantitative imaging biomarkers. These concepts are presented first as examples of conversations between investigator and biostatistician, and then more detailed discussions of the statistical concepts follow. Three skeletal radiology examples are used to illustrate the concepts.

Keywords

noninferiority study - interchangeability - multiplicity - confidence intervals - quantitative imaging biomarkers

• References

• 1 Subhas N, Purysko CP, Obuchowski NA. , et al. Using a dedicated CT metal artifact reduction algorithm for arthroplasty imaging: can we reduce the dose? Paper presented at: Society of Skeletal Radiology 39th Annual Meeting; March 13–16, 2016; New Orleans, LA
• 2 Subhas N, Benedick A, Obuchowski NA. , et al. 5-minute shoulder MRI: is it good enough? Paper presented at: Society of Skeletal Radiology 39th Annual Meeting; March 13–16, 2016 ; New Orleans, LA
• 3 Polster JM, Bullen J, Obuchowski NA, Bryan JA, Soloff L, Schickendantz MS. Relationship between humeral torsion and injury in professional baseball pitchers. Am J Sports Med 2013; 41 (09) 2015-2021
• 4 Chen W, Petrick NA, Sahiner B. Hypothesis testing in noninferiority and equivalence MRMC ROC studies. Acad Radiol 2012; 19 (09) 1158-1165
• 5 Zhou XH, Obuchowski NA, McClish DK. . Statistical Methods in Diagnostic Medicine. Hoboken, NJ; Wiley and Sons; 2011
• 6 Obuchowski NA, Subhas N, Schoenhagen P. Testing for interchangeability of imaging tests. Acad Radiol 2014; 21 (11) 1483-1489
• 7 Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat 1979; 6: 65-70
• 8 Dmitrienko A, Bretz F, Westfall PH. , et al. Multiple testing methodology. In: Dmitrienko A, Tamhane AC, Bretz F, eds. Multiple Testing Problems in Pharmaceutical Statistics. Boca Raton, FL: Chapman & Hall/CRC Biostatistics Series; 2009
• 9 Westfall PH, Krishen A. Optimally weighted, fixed sequence, and gatekeeping multiple testing procedures. J Stat Plan Inference 2001; 99: 25-40
• 10 Dunn OJ. Multiple comparisons among means. J Am Stat Assoc 1961; 56: 52-64
• 11 Sullivan DC, Obuchowski NA, Kessler LG. , et al; RSNA-QIBA Metrology Working Group. Metrology standards for quantitative imaging biomarkers. Radiology 2015; 277 (03) 813-825
• 12 Kessler LG, Barnhart HX, Buckler AJ. , et al; QIBA Terminology Working Group. The emerging science of quantitative imaging biomarkers terminology and definitions for scientific studies and regulatory submissions. Stat Methods Med Res 2015; 24 (01) 9-26
• 13 Polster JM, Subhas N, Scalise JJ, Bryan JA, Lieber ML, Schickendantz MS. Three-dimensional volume-rendering computed tomography for measuring humeral version. J Shoulder Elbow Surg 2010; 19 (06) 899-907
• 14 Raunig DL, McShane LM, Pennello G. , et al; QIBA Technical Performance Working Group. Quantitative imaging biomarkers: a review of statistical methods for technical performance assessment. Stat Methods Med Res 2015; 24 (01) 27-67
• 15 Obuchowski NA, Reeves AP, Huang EP. , et al; Algorithm Comparison Working Group. Quantitative imaging biomarkers: a review of statistical methods for computer algorithm comparisons. Stat Methods Med Res 2015; 24 (01) 68-106
• 16 Obuchowski NA, Barnhart HX, Buckler AJ. , et al; Case Example Working Group. Statistical issues in the comparison of quantitative imaging biomarker algorithms using pulmonary nodule volume as an example. Stat Methods Med Res 2015; 24 (01) 107-140