Logo
RadAI Slice
Subscribe

Automated Detection of Focal Bone Marrow Lesions From MRI: A Multi-center Feasibility Study in Patients with Monoclonal Plasma Cell Disorders.

July 9, 2025pubmed logopapersDOI: 10.1016/j.acra.2025.06.034PMID: 40640054

Authors

Wennmann M,Kächele J,von Salomon A,Nonnenmacher T,Bujotzek M,Xiao S,Martinez Mora A,Hielscher T,Hajiyianni M,Menis E,Grözinger M,Bauer F,Riebl V,Rotkopf LT,Zhang KS,Afat S,Besemer B,Hoffmann M,Ringelstein A,Graeven U,Fedders D,Hänel M,Antoch G,Fenk R,Mahnken AH,Mann C,Mokry T,Raab MS,Weinhold N,Mai EK,Goldschmidt H,Weber TF,Delorme S,Neher P,Schlemmer HP,Maier-Hein K

Affiliations (23)

  • Division of Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (M.W., A.V.S., M.G., F.B., L.T.R., K.S.Z., S.D., H.P.S.); Diagnostic and Interventional Radiology, Heidelberg University Hospital, Im Neuenheimer Feld 410, 69120 Heidelberg, Germany (M.W., T.N., V.R., T.M., T.F.W.). Electronic address: [email protected].
  • Division of Medical Image Computing, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (J.K., M.B., S.X., A.M.M., P.N., K.M.H.); Medical Faculty, Heidelberg University, Im Neuenheimer Feld 672, 69120 Heidelberg, Germany (J.K., A.V.S., M.B., A.M.M.).
  • Division of Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (M.W., A.V.S., M.G., F.B., L.T.R., K.S.Z., S.D., H.P.S.); Medical Faculty, Heidelberg University, Im Neuenheimer Feld 672, 69120 Heidelberg, Germany (J.K., A.V.S., M.B., A.M.M.).
  • Diagnostic and Interventional Radiology, Heidelberg University Hospital, Im Neuenheimer Feld 410, 69120 Heidelberg, Germany (M.W., T.N., V.R., T.M., T.F.W.).
  • Division of Medical Image Computing, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (J.K., M.B., S.X., A.M.M., P.N., K.M.H.); Faculty of Mathematics and Computer Science, Heidelberg University, Im Neuenheimer Feld 205, 69120 Heidelberg, Germany (S.X.).
  • Division of Biostatistics, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (T.H.).
  • Department of Internal Medicine V, Heidelberg Myeloma Center, Heidelberg University Hospital, Im Neuenheimer Feld 410, 69120 Heidelberg, Germany (M.H., M.S.R., N.W.).
  • Internal Medicine V, Hematology, Oncology and Rheumatology, GMMG Study Group, Heidelberg University Hospital, Im Neuenheimer Feld 130.3, 69120 Heidelberg, Germany (E.M., E.K.M., H.G.).
  • Division of Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (M.W., A.V.S., M.G., F.B., L.T.R., K.S.Z., S.D., H.P.S.).
  • Division of Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (M.W., A.V.S., M.G., F.B., L.T.R., K.S.Z., S.D., H.P.S.); Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Kerpener Str. 62/Gebäude 18a, 50937 Cologne, Germany (F.B.).
  • Department of Radiology, Diagnostic and Interventional Radiology, University Hospital of Tuebingen, Hoppe-Seyler-Straße 3, 72076 Tübingen, Germany (S.A.).
  • Department of Hematology, Oncology, Immunology and Rheumatology, University Hospital of Tuebingen, Otfried-Müller-Straße 10, 72076 Tübingen, Germany (B.B.).
  • Medical Clinic A, Hospital of Ludwigshafen, Bremserstraße 79, 67063 Ludwigshafen am Rhein, Germany (M.H.).
  • Department of Radiology and Neuroradiology, Kliniken Maria Hilf GmbH, Viersener Str. 450, 41063 Mönchengladbach, Germany (A.R.).
  • Department of Hematology, Oncology, and Gastroenterology, Kliniken Maria Hilf GmbH, Viersener Str. 450, 41063 Mönchengladbach, Germany (U.G.).
  • Institute for Radiology and Neuroradiology, Clinic Chemnitz, Flemmingstraße 2, 09116 Chemnitz, Germany (D.F.).
  • Department of Internal Medicine III, Clinic Chemnitz, Flemmingstraße 2, 09116 Chemnitz, Germany (M.H.).
  • University Dusseldorf, Medical Faculty, Department of Diagnostic and Interventional Radiology, Moorenstrasse 5, 40225 Düsseldorf, Germany (G.A.); Center for Integrated Oncology (CIO Aachen, Bonn, Cologne, Duesseldorf), Moorenstraße 5, 40225 Düsseldorf, Germany (G.A.).
  • Department of Hematology, Oncology, and Clinical Immunology, University Hospital Düsseldorf, Moorenstraße 5, 40225 Düsseldorf, Germany (R.F.).
  • Clinic of Diagnostic and Interventional Radiology, Marburg University Hospital, Philipps-University Marburg, Baldingerstrasse, 35043 Marburg, Germany (A.H.M.).
  • Department for Hematology, Oncology, and Immunology, University Hospital of Gießen and Marburg, Baldingerstrasse, 35043 Marburg, Germany (C.M.).
  • Internal Medicine V, Hematology, Oncology and Rheumatology, GMMG Study Group, Heidelberg University Hospital, Im Neuenheimer Feld 130.3, 69120 Heidelberg, Germany (E.M., E.K.M., H.G.); National Center for Tumor Diseases (NCT) Heidelberg, Im Neuenheimer Feld 460, 69120 Heidelberg, Germany (H.G.).
  • Division of Medical Image Computing, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (J.K., M.B., S.X., A.M.M., P.N., K.M.H.); Pattern Analysis and Learning Group, Department of Radiation Oncology, Heidelberg University Hospital, Im Neuenheimer Feld 410, 69120 Heidelberg, Germany (P.N., K.M.H.); German Cancer Consortium (DKTK), Partner Site Heidelberg, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany (P.N., K.M.H.).

Abstract

To train and test an AI-based algorithm for automated detection of focal bone marrow lesions (FL) from MRI. This retrospective feasibility study included 444 patients with monoclonal plasma cell disorders. For this feasibility study, only FLs in the left pelvis were included. Using the nnDetection framework, the algorithm was trained based on 334 patients with 494 FLs from center 1, and was tested on an internal test set (36 patients, 89 FLs, center 1) and a multicentric external test set (74 patients, 262 FLs, centers 2-11). Mean average precision (mAP), F1-score, sensitivity, positive predictive value (PPV), and Spearman correlation coefficient between automatically determined and actual number of FLs were calculated. On the internal/external test set, the algorithm achieved a mAP of 0.44/0.34, F1-Score of 0.54/0.44, sensitivity of 0.49/0.34, and a PPV of 0.61/0.61, respectively. In two subsets of the external multicentric test set with high imaging quality, the performance nearly matched that of the internal test set, with mAP of 0.45/0.41, F1-Score of 0.50/0.53, sensitivity of 0.44/0.43, and a PPV of 0.60/0.71, respectively. There was a significant correlation between the automatically determined and actual number of FLs on both the internal (r=0.51, p=0.001) and external multicentric test set (r=0.59, p<0.001). This study demonstrates that the automated detection of FLs from MRI, and thereby the automated assessment of the number of FLs, is feasible.

View SourceFull Text PDF

Tags

MRIDetectionMusculoskeletalRetrospective ClinicalIn SilicoAcademic Lab

Topics

Journal Article

Ready to Sharpen Your Edge?

Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.