Ideas for Dealing with Reduced Datasets in Development of CADe Systems for Medical Uses
Resumo
This work presents a real time user friendly system to aid specialized professionals to analyze bone scans exams. In order to achieve this, some original ideas are applied. The first one is related to the use of each pixel of an exam as object of interest for classification. Another original idea is the use of operations that are normally applied in pre-processing as features for machine learning. With both, even using small dataset was possible to obtain enough amounts of entries to be used for training and testing. Initially, the feature vectors are composed by 64 features and one target attribute representing the classification result. The used bone scans set was composed of 42 images from 21 patients. At the end of the learning tasks a dataset of 2,512,386 records is computed. In order to reduce the cardinality of the vector of features, the Principal Component Analysis was employed leading to a new feature set with 25 components per object to be classified as with or without metastasis, the area under the Receiver Operator Characteristic curve achieved with this final set of features was 98%.
Referências
Wyant, T. Bone metastases. Cancer.org. 2021 Available at: [link]. Accessed: at 28 July 2022.
Wyngaert, T.; Strobel, K.; Kampen, W.; Kuwert, T.; Bruggen, W.; Mohan, H.; Gnanasegaran, G.; Delgado-Bolton, R.; Weber, W.; Beheshti, W.; Langsteger, M.; Giammarile, W.; Mottaghy, F.; Paycha, F. The EANM practice guidelines for bone scintigraphy. European Journal of Nuclear Medicine and Molecular Imaging. 2016
Rieden, K. Conventional imaging and computerized tomography in diagnosis of skeletal metastases Radiologe. Der Radiologe. 1995.
Westmead BCI Institute. Bone Metastasis. Westmead BCI Institute. 2017. Available at: [link]. Accessed at: 20 July 2022.
Łukaszewski, B.; Nazar, J.; Goch, M.; Łukaszewska, M.; Stepiński, A.; Jurczyk, M. Diagnostic methods for detection of bone metastases. Contemporary oncology (Poznan, Poland), 21(2), 98-103. https://doi.org/10.5114/wo.2017.68617. 2017.
Donohoe K.; Brown M.; Collier, D.; Carretta, R.; Henkin, R.; O'mara, R.; Royal, H. Society of Nuclear Medicine Procedure Guideline for Bone Scintigraphy. 2023.
Hopkins Medicine. Bone Scan. Hopkins Medicine. 2021. Available at: [link]. Accessed at: 10 July. 2022
Erdi, Y.E.; Humm, J.L.; Imbriaco, M.; Yeung, H.; Larson, S.M. Quantitative bone metastases analysis based on image segmentation. J. Nucl. Med. 38: 1401-6. 1997
Ulmert, D.; Kaboteh, R.; Fox J.J. A novel automated platform for quantifying the extent of skeletal tumour involvement in prostate cancer patients using the Bone Scan Index. Eur Urol. 62(1):78-84.PubMedGoogle ScholarCrossref. 2012.
Sadik, M. Jakobsson, D.; Olofsson, F.; Ohlsson, M.; Suurkula, M.; Edenbrandt, L. A new computer-based decision-support system for the interpreta tion of bone scans. Nuclear medicine communications. 2006.
Sadik, M. Hamadeh, I.; Nordblom, P.; Suurkula, M.; Hoglund, P.; Ogl Soon, M.;Edenbrandt, L. Computer-Assisted Interpretation of Planar Whole-Body Bone Scans. J Nucl Med. 2008.
Nakajima, K; Edenbrandt L; Mizokami, A. Bone scan index: A new biomarker of bone metastasis in patients with prostate cancer.Int J Urol. 2017.
Horikoshi, H.; Kikuchi A.; Onoguchi M. Computeraided diagnosis system for bone scintigrams from Japanese patients:importance of training database. Ann Nuclear Medicine. 2012
Nakajima, K.; Nakajima, Y.; Horikoshi, H. Enhanced diagnostic accuracy for quantitative bone scan using an artificial neural network system: a Japanese multi-center database project. EJNMMI Res. 2013.
Aslantas, A.; Dandil, E.; Saglam, S.; Çakiroglu, M. Cadboss: A computer-aided diagnosis system for whole-body bone scintigraphy scans. Journal of cancer research and therapeutics, 12(2), 787. 2016.
Dang, J. Classification in bone scintigraphy images using convolutional neural networks. Master's Theses in Mathematical Sciences. Lund University, Sweden, 2016.
Belcher, L. Convolutional neural networks for classification of prostate cancer metastases using bone scan images. Theoretical Physics Masters Project, Lund University. Sweden , 2017
Papandrianos N.; Papageorgiou E.; Anagnostis A.; Papageorgiou K. Bone metastasis classification using whole body images from prostate cancer patients based on convolutional neural networks application. PLoS ONE. 2020.
APEER, An intuitive annotation tool for your machine learning needs. APEER. Available at: https://www.apeer.com/annotate. Accessed at: 30 dec. 2021.
M. Sonka, V. Hlavac and R. Boyle, Image Processing, Analysis and Machine Vision, 3th Edition, Thomson, 2008.
Rosebrock, A. OpenCV Template Matching. Pyimagesearch. 2021 Available at: [link]. Accessed at: 20 July. 2022.
Scikit image. Entropy. Scikit image. 2021. Available at: [link]. Accessed at: 20 July 2022.
Litjens, G., Kooi, T., Bejnordi, B.E., Setio, A.A.A., Ciompi, F., Ghafoorian, M., van der Laak, J.A., Van Ginneken, B., Sánchez, C.I.: A survey on deep learning in medical image analysis. Med. Image Anal. 42, 60-88 (2017).
Haskins, G., Kruger, U. & Yan, P. Deep learning in medical image registration: a survey. Machine Vision and Applications 31, 8 (2020). https://doi.org/10.1007/s00138-020-01060-x
Myronenko, A., Song, X.: Intensity-based image registration by minimizing residual complexity. IEEE Trans. Med. Imaging 29(11), 1882-1891 (2010).
Carneiro da Silva, J. M., Bone-scan: a Computer-aided Detection (CADe) and Diagnosis (CADx) System for Bone Scintigraphy, Compter Graduation Program, Master's Dissertation, Computer Institute, UFF, 2022 [link]