Advertisement

Random forest classifiers aid in the detection of incidental osteoblastic osseous metastases in DEXA studies

  • Samir D. Mehta
  • Ronnie SebroEmail author
Short communication
  • 40 Downloads

Abstract

Purpose

Dual-energy X-ray absorptiometry (DEXA) studies are used for screening patients for low bone mineral density (BMD). Patients with breast and prostate cancer are often treated with hormone-altering drugs that result in low BMD. These patients may have incidental osteoblastic metastases of the spine that may be detected on screening DEXA studies. The aim of this pilot study is to assess whether random forest classifiers or support vector machines can identify patients with incidental osteoblastic metastases of the spine from screening DEXA studies and to evaluate which technique is better.

Methods

We retrospectively reviewed the DEXA studies from 200 patients (155 normal control patients and 45 patients with osteoblastic metastases of one or more lumbar vertebral bodies from L1 to L4). The dataset was split into training (80%) and validation (20%) datasets. The optimal random forest (RF) and support vector machine (SVM) classifiers were obtained. Receiver-operator-characteristic curves were compared with DeLong’s test.

Results

The sensitivity, specificity, accuracy and area under the curve (AUC) of the optimal RF classifier were 77.8%, 100.0%, 98.0% and 0.889, respectively, in the validation dataset. The sensitivity, specificity, accuracy and AUC of the optimal SVM classifier were 33.3%, 96.8%, 82.5% and 0.651 in the validation dataset. The RF classifier was significantly better than the SVM classifier (P = 0.008). Only 7 of the 45 patients with osteoblastic metastases (15.6%) were prospectively identified by the radiologist interpreting the study.

Conclusions

RF classifiers can be used as a useful adjunct to identify incidental lumbar spine osteoblastic metastases in screening DEXA studies.

Keywords

Random forest Support vector machine DEXA, machine learning Artificial intelligence 

Notes

Funding

RS was funded by a Radiology Society of North America Research Grant.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Human and animal rights statement

All human studies have been approved and performed in accordance with ethical standards.

Informed consent

The study was reviewed and approved by the senior author’s Institutional Review Board, and the need for signed informed consent from each participant was waived.

References

  1. 1.
    Rachner TD, Khosla S, Hofbauer LC (2011) Osteoporosis: now and the future. Lancet 377(9773):1276–1287.  https://doi.org/10.1016/S0140-6736(10)62349-5 CrossRefGoogle Scholar
  2. 2.
    Eriksen EF (2012) Treatment of osteopenia. Rev Endocr Metab Disord 13(3):209–223.  https://doi.org/10.1007/s11154-011-9187-z CrossRefGoogle Scholar
  3. 3.
    Qaseem A, Forciea MA, McLean RM, Denberg TD, Clinical Guidelines Committee of the American College of Physicians (2017) Treatment of low bone density or osteoporosis to prevent fractures in men and women: a clinical practice guideline update from the American College of Physicians. Ann Intern Med 166(11):818–839CrossRefGoogle Scholar
  4. 4.
    Golob AL, Laya MB (2015) Osteoporosis: screening, prevention, and management. Med Clin N. Am 99(3):587–606CrossRefGoogle Scholar
  5. 5.
    Colón-Emeric CS, Saag KG (2006) Osteoporotic fractures in older adults. Best Pract Res Clin Rheumatol 20(4):695–706CrossRefGoogle Scholar
  6. 6.
    Lewiecki EM (2013) Bone density measurement and assessment of fracture risk. Clin Obstet Gynecol 56(4):667–676CrossRefGoogle Scholar
  7. 7.
    U.S. Preventive Services Task Force (2011) Screening for osteoporosis: US preventive services task force recommendation statement. Ann Intern Med 154(5):356–364CrossRefGoogle Scholar
  8. 8.
    WHO Scientific Group on the Prevention and Management of Osteoporosis (2000: Geneva, Switzerland) (2003). Prevention and management of osteoporosis: report of a WHO scientific groupGoogle Scholar
  9. 9.
    Mazess RB, Barden HS (1988) Measurement of bone by dual-photon absorptiometry (DPA) and dual-energy X-ray absorptiometry (DEXA). Ann Chir Gynaecol 77(5–6):197–203Google Scholar
  10. 10.
    Syed Z, Khan A (2002) Bone densitometry: applications and limitations. J Obstet Gynaecol Can 24(6):476–484CrossRefGoogle Scholar
  11. 11.
    Watts NB (2004) Fundamentals and pitfalls of bone densitometry using dual-energy X-ray absorptiometry (DXA). Osteoporos Int 15(11):847–854CrossRefGoogle Scholar
  12. 12.
    Blake GM, Fogelman I (2007) The role of DXA bone density scans in the diagnosis and treatment of osteoporosis. Postgrad Med J 83(982):509–517CrossRefGoogle Scholar
  13. 13.
    Kanis JA on behalf of the World Health Organization Scientific Group (2007) Assessment of osteoporosis at the primary health-care level. Technical Report. University of Sheffield, UK World Health Organization Collaborating Centre for Metabolic Bone Diseases. University of Sheffield, 2007Google Scholar
  14. 14.
    Wright NC, Looker AC, Saag KG, Curtis JR, Delzell ES, Randall S, Dawson-Hughes B (2014) The recent prevalence of osteoporosis and low bone mass in the United States based on bone mineral density at the femoral neck or lumbar spine. J Bone Miner Res 29(11):2520–2526.  https://doi.org/10.1002/jbmr.2269 CrossRefGoogle Scholar
  15. 15.
    Eastell R, Hannon RA, Cuzick J, Dowsett M, Clack G, Adams JE, ATAC Trialists group (2006) Effect of an aromatase inhibitor on bmd and bone turnover markers: 2-year results of the Anastrozole, Tamoxifen, Alone or in Combination (ATAC) trial (18233230). J Bone Miner Res 21(8):1215–1223CrossRefGoogle Scholar
  16. 16.
    Mincey BA, Duh MS, Thomas SK, Moyneur E, Marynchencko M, Boyce SP, Mallett D, Perez EA (2006) Risk of cancer treatment-associated bone loss and fractures among women with breast cancer receiving aromatase inhibitors. Clin Breast Cancer 7(2):127–132CrossRefGoogle Scholar
  17. 17.
    Hillner BE, Ingle JN, Chlebowski RT, Gralow J, Yee GC, Janjan NA, Cauley JA, Blumenstein BA, Albain KS, Lipton A, Brown S, American Society of Clinical Oncology (2003) American Society of Clinical Oncology 2003 update on the role of bisphosphonates and bone health issues in women with breast cancer. J Clin Oncol 21(21):4042–4057 Epub 2003 Sep 8 CrossRefGoogle Scholar
  18. 18.
    Diamond TH, Higano CS, Smith MR, Guise TA, Singer FR (2004) Osteoporosis in men with prostate carcinoma receiving androgen-deprivation therapy: recommendations for diagnosis and therapies. Cancer 100:892–899CrossRefGoogle Scholar
  19. 19.
    Smith MR (2003) Diagnosis and management of treatment-related osteoporosis in men with prostate carcinoma. Cancer 97(3 Suppl):789–795CrossRefGoogle Scholar
  20. 20.
    National Institute for Health and Clinical Excellence (2018) Early and locally advanced breast cancer: Diagnosis and treatment (NG101)Google Scholar
  21. 21.
    Bazzocchi A, Ferrari F, Diano D, Albisinni U, Battista G, Rossi C, Guglielmi G (2012) Incidental findings with dual-energy X-ray absorptiometry: spectrum of possible diagnoses. Calcif Tissue Int 91(2):149–156CrossRefGoogle Scholar
  22. 22.
    Kahn CE Jr (2017) From images to actions: opportunities for artificial intelligence in radiology. Radiology 285(3):719–720CrossRefGoogle Scholar
  23. 23.
    Kahn CE Jr (1994) Artificial intelligence in radiology: decision support systems. Radiographics 14(4):849–861CrossRefGoogle Scholar
  24. 24.
    Wu Y, Giger ML, Doi K, Vyborny CJ, Schmidt RA, Metz CE (1993) Artificial neural networks in mammography: application to decision making in the diagnosis of breast cancer. Radiology 187(1):81–87CrossRefGoogle Scholar
  25. 25.
    Sarica A, Cerasa A, Quattrone A (2017) Random forest algorithm for the classification of neuroimaging data in Alzheimer’s disease: a systematic review. Front Aging Neurosci 9:329.  https://doi.org/10.3389/fnagi.2017.00329 eCollection 2017 CrossRefGoogle Scholar
  26. 26.
    Kruppa J, Liu Y, Biau G, Kohler M, König IR, Malley JD, Ziegler A (2014) Probability estimation with machine learning methods for dichotomous and multicategory outcome: theory. Biom J 56(4):534–563CrossRefGoogle Scholar
  27. 27.
    Ben-Hur A, Horn D, Siegelmann H, Vapnik V (2001) Support vector clustering. J Mach Learn Res 2:125–137Google Scholar
  28. 28.
    DeLong ER, DeLong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44(3):837–845CrossRefGoogle Scholar
  29. 29.
    Forbes V, Taxel P (2018) Onset of asymptomatic skeletal metastatic disease seen on DXA. AACE Clin Case Rep 2018(4):e472–e475.  https://doi.org/10.4158/ACCR-2017-0163 CrossRefGoogle Scholar
  30. 30.
    Martineau P, Bazarjani S, Zuckier LS (2015) Artifacts and incidental findings encountered on dual-energy X-ray absorptiometry: atlas and analysis. Semin Nucl Med 45(5):458–469.  https://doi.org/10.1053/j.semnuclmed.2015.02.001 CrossRefGoogle Scholar

Copyright information

© CARS 2019

Authors and Affiliations

  1. 1.Department of RadiologyUniversity of PennsylvaniaPhiladelphiaUSA
  2. 2.Department of Orthopedic SurgeryUniversity of PennsylvaniaPhiladelphiaUSA
  3. 3.Department of GeneticsUniversity of PennsylvaniaPhiladelphiaUSA
  4. 4.Department of Epidemiology and BiostatisticsUniversity of PennsylvaniaPhiladelphiaUSA

Personalised recommendations