Abstract
Despite the existence of patterns able to discriminate between consensual and non-consensual intercourse, the relevance of genital lesions in the corroboration of a legal rape complaint is currently under debate in many countries. The testimony of the physicians when assessing these lesions has been questioned in court due to several factors (e.g., a lack of comprehensive knowledge of lesions, wide spectrum of background area, among others). Therefore, it is relevant to provide automated tools to support the decision process in an objective manner. In this work, we evaluate the performance of state-of-the-art deep learning architectures for the forensic assessment of sexual assault. We propose a deep architecture and learning strategy to tackle the class imbalance on deep learning using ranking. The proposed methodologies achieved the best results when compared with handcrafted feature engineering and with other deep architectures .
Similar content being viewed by others
References
Astrup BS, Lauritsen J, Ravn P, Thomsen JL (2011) Genital lesions after consensual sexual intercourse: they are frequent and they last for several days. In: 19th world meeting of the international association of forensic sciences
Astrup BS, Lauritsen J, Thomsen JL, Ravn P (2013) Colposcopic photography of genital injury following sexual intercourse in adults. Forensic Sci Med Pathol 9(1):24–30
Astrup BS, Lykkebo AW (2015) Post-coital genital injury in healthy women: a review. Clin Anat 28(3):331–338
Astrup BS, Ravn P, Lauritsen J, Thomsen JL (2012) Nature, frequency and duration of intercourseimplications for legal proceedings. Forensic Sci Int 219(1):50–56
Astrup BS, Ravn P, Thomsen JL, Lauritsen J (2013) Patterned genital injury in cases of rape-a case–control study. J Forensic Legal Med 20(5):525–529
Avila S, Thome N, Cord M, Valle E, AraúJo ADA (2013) Pooling in image representation: the visual codeword point of view. Comput Vis Image Underst 117(5):453–465
Cruz R, Fernandes K, Cardoso JS, Costa JFP (2016) Tackling class imbalance with ranking. In: 2016 International joint conference on neural networks (IJCNN), pp 2182–2187. IEEE
Cruz R, Fernandes K, Costa JFP, Cardoso JS (2017) Constraining type ii error: building intentionally biased classifiers. In: International work-conference on artificial neural networks, pp 549–560. Springer
Cruz R, Fernandes K, Costa JFP, Ortiz MP, Cardoso JS (2017) Combining ranking with traditional methods for ordinal class imbalance. In: International work-conference on artificial neural networks, pp 538–548. Springer
Cruz R, Fernandes K, Costa JFP, Ortiz MP, Cardoso JS (2017) Ordinal class imbalance with ranking. In: Iberian conference on pattern recognition and image analysis, pp 3–12. Springer
Erhan D, Bengio Y, Courville A, Manzagol PA, Vincent P, Bengio S (2010) Why does unsupervised pre-training help deep learning? J Mach Learn Res 11(Feb):625–660
Fernandes K, Cardoso JS, Astrup BS (2017) Automated detection and categorization of genital injuries using digital colposcopy. In: Iberian conference on pattern recognition and image analysis, pp 251–258. Springer
Fernandes K, Cardoso JS, Fernandes J (2015) Temporal segmentation of digital colposcopies. In: Iberian conference on pattern recognition and image analysis, pp 262–271. Springer
Fernandes K, Cardoso JS, Fernandes J (2017) Transfer learning with partial observability applied to cervical cancer screening. In: Iberian conference on pattern recognition and image analysis, pp 243–250. Springer
Glorot X, Bordes A, Bengio Y (2011) Deep sparse rectifier neural networks. In: Proceedings of the fourteenth international conference on artificial intelligence and statistics, pp 315–323
González-Castro V, Alaiz-Rodríguez R, Fernández-Robles L, Guzmán-Martínez R, Alegre E (2010) Estimating class proportions in boar semen analysis using the hellinger distance. In: García-Pedrajas N, Herrera F, Fyfe C, Benítez JM, Ali M (eds) Trends in Applied Intelligent Systems. IEA/AIE 2010. Lecture Notes in Computer Science, vol 6096. Springer, Berlin, Heidelberg, pp 284–293
Goodfellow IJ, Shlens J, Szegedy C (2014) Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778
Huang X, Wang W, Xue Z, Antani S, Long LR, Jeronimo J (2008) Tissue classification using cluster features for lesion detection in digital cervigrams. In: Medical imaging, pp 69,141Z–69,141Z. International Society for Optics and Photonics
Lee H, Pham P, Largman Y, Ng AY (2009) Unsupervised feature learning for audio classification using convolutional deep belief networks. In: Advances in neural information processing systems, pp 1096–1104
Long J, Shelhamer E, Darrell T (2015) Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 3431–3440
Lopes APB, de Avila SE, Peixoto AN, Oliveira RS, Coelho MdM, Araújo AdA (2009) Nude detection in video using bag-of-visual-features. In: 2009 XXII Brazilian symposium on computer graphics and image processing (SIBGRAPI), pp 224–231. IEEE
Lopes APB, de Avila SEF, Peixoto ANA, Oliveira RS, de Albuquerque Araújo A (2009) A bag-of-features approach based on hue-sift descriptor for nude detection. In: Proceedings of the XVII European signal processing conference (EUSIPCO), Glasgow, Scotland
Papernot N, McDaniel P, Goodfellow I, Jha S, Celik ZB, Swami A (2017) Practical black-box attacks against machine learning. In: Proceedings of the 2017 ACM on Asia conference on computer and communications security, pp 506–519. ACM
Papernot N, McDaniel P, Jha S, Fredrikson M, Celik ZB, Swami A (2016) The limitations of deep learning in adversarial settings. In: 2016 IEEE European symposium on security and privacy (EuroS&P), pp 372–387. IEEE
Payne-James J, Busuttil A, Smock W (2003) Forensic medicine: clinical and pathological aspects. Cambridge University Press, Cambridge
Pérez-Ortiz M, Fernandes K, Cruz R, Cardoso JS, Briceño J, Hervás-Martínez C (2017) Fine-to-coarse ranking in ordinal and imbalanced domains: an application to liver transplantation. In: International work-conference on artificial neural networks, pp 525–537. Springer
Ronneberger O, Fischer P, Brox T (2015) U-net: Convolutional networks for biomedical image segmentation. In: International conference on medical image computing and computer-assisted intervention, pp 234–241. Springer
Rosasco L, Tacchetti A, Villa S (2014) Regularization by early stopping for online learning algorithms. Stat 1050:30
Slaughter L, Brown CR, Crowley S, Peck R (1997) Patterns of genital injury in female sexual assault victims. Am J Obstet Gynecol 176(3):609–616
Srivastava N, Hinton GE, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958
Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z (2016) Rethinking the inception architecture for computer vision. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2818–2826
Yao Y, Rosasco L, Caponnetto A (2007) On early stopping in gradient descent learning. Constr Approx 26(2):289–315
Zeiler MD (2012) Adadelta: an adaptive learning rate method. arXiv preprint arXiv:1212.5701
Zeiler MD, Fergus R (2014) Visualizing and understanding convolutional networks. In: European conference on computer vision, pp 818–833. Springer
Zhou B, Lapedriza A, Xiao J, Torralba A, Oliva A (2014) Learning deep features for scene recognition using places database. In: Advances in neural information processing systems, pp 487–495
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Fernandes, K., Cardoso, J.S. & Astrup, B.S. A deep learning approach for the forensic evaluation of sexual assault. Pattern Anal Applic 21, 629–640 (2018). https://doi.org/10.1007/s10044-018-0694-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10044-018-0694-3