Samir D Mehta1, Ronnie Sebro2,3,4,5. 1. Department of Radiology, University of Pennsylvania, 3737 Market Street, Philadelphia, PA, 19104, USA. 2. Department of Radiology, University of Pennsylvania, 3737 Market Street, Philadelphia, PA, 19104, USA. Ronnie.sebro@uphs.upenn.edu. 3. Department of Orthopedic Surgery, University of Pennsylvania, 3737 Market Street, Philadelphia, PA, 19104, USA. Ronnie.sebro@uphs.upenn.edu. 4. Department of Genetics, University of Pennsylvania, 421 Marie Curie Blvd, Philadelphia, PA, 19104, USA. Ronnie.sebro@uphs.upenn.edu. 5. Department of Epidemiology and Biostatistics, University of Pennsylvania, 421 Marie Curie Blvd, Philadelphia, PA, 19104, USA. Ronnie.sebro@uphs.upenn.edu.
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.
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.
Entities:
Keywords:
Artificial intelligence; DEXA, machine learning; Random forest; Support vector machine
Authors: Betty A Mincey; Mei Sheng Duh; Simu K Thomas; Erick Moyneur; Maryna Marynchencko; Simone Peart Boyce; David Mallett; Edith A Perez Journal: Clin Breast Cancer Date: 2006-06 Impact factor: 3.225
Authors: Richard Eastell; Rosemary A Hannon; Jack Cuzick; Mitch Dowsett; Glen Clack; Judith E Adams Journal: J Bone Miner Res Date: 2006-08 Impact factor: 6.741
Authors: Bruce E Hillner; James N Ingle; Rowan T Chlebowski; Julie Gralow; Gary C Yee; Nora A Janjan; Jane A Cauley; Brent A Blumenstein; Kathy S Albain; Allan Lipton; Susan Brown Journal: J Clin Oncol Date: 2003-09-08 Impact factor: 44.544
Authors: Terrence H Diamond; Celestia S Higano; Matthew R Smith; Theresa A Guise; Frederick R Singer Journal: Cancer Date: 2004-03-01 Impact factor: 6.860
Authors: Yangyang Cui; Jia Zhu; Zhili Duan; Zhenhua Liao; Song Wang; Weiqiang Liu Journal: Int J Environ Res Public Health Date: 2022-09-16 Impact factor: 4.614