Abstract
医療分野において，完全なアノテーションが与えられた学習データを収集することは難しい．本論文では腹部超音波画像からの肝腫瘍検出を対象に，画像内の一部の腫瘍に位置情報を与えた，部分的にアノテーションされた転移性肝がんの検出についてYOLOv3の損失関数を調節する手法を提案する．YOLOv3の損失関数の背景領域に関する項に対して腫瘍の種類に応じた重みを加えることで，アノテーションがない腫瘍に対する検出精度の改善を試みた．転移性肝がんの重みのみ小さくした実験の結果，転移性肝がんに対する再現率の向上が確認された．また，重み0.1のとき，アノテーションが与えられた転移性肝がんの適合率は1割程度低下したが，これは評価データにおいてアノテーションが与えられていない腫瘍を検出したためである．他の腫瘍においては再現率，適合率ともに精度に変化はなかった．この結果から，本手法の有効性が示唆された．
Translated Abstract
Collecting accurately annotated training data in the medical field is challenging. This paper proposes a method to adapt the YOLOv3 loss function for detecting metastatic liver cancer from abdominal ultrasound images, even with incomplete annotations. We aim to enhance detection accuracy for unannotated tumors by incorporating tumor-type-specific weights into the background component of the YOLOv3 loss function. After conducting experiments where only the weight for metastatic liver cancer was reduced, we observed improved recall rates for this tumor type. Furthermore, when the weight was set to 0.1, the precision for annotated metastatic liver cancer decreased by approximately 10％, due to the detection of non-annotated tumors during evaluation. It is important to note that the detection accuracy for other tumor types did not diminish. These findings indicate the effectiveness of our proposed method.
―腹部超音波画像からの肝腫瘍検出における評価―
Ultrasound image, Detection of liver tumor, Deep learning, Incomplete Annotation
References
- ［ 1 ］ Girshick R, Donahue J, Darrell T, et al.: Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, 2014, pp. 580-587
- ［ 2 ］ Carion N, Massa F, Synnaeve G, et al.: End-to-end object detection with transformers. In Proceedings of the 16th European Conference on Computer Vision (ECCV), Glasgow, 2020, pp. 213-229
- ［ 3 ］ Liu W, Anguelov D, Erhan D, et al.: SSD: Single Shot MultiBox Detector. In Proceedings of the 14th European Conference on Computer Vision（ ECCV）, Amsterdam, 2016, LNCS Vol. 9905, Springer, pp. 21-37
- ［ 4 ］ Zhou X, Wang D, Krähenbühl P: Objects as points. arXiv preprint arXiv:1904.07850, 2019
- ［ 5 ］ Zhu X, Hu H, Lin S, et al.: Deformable ConvNets v2: More deformable, better results. In Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, 2019, pp. 9308-9316
- ［ 6 ］ Redmon J, Divvala S, Girshick R, et al.: You only look once: Unified, real-time object detection. In Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 2016, pp. 779-788
- ［ 7 ］ Li K, Fathan MI, Patel K, et al.: Colonoscopy polyp detection and classification: Dataset creation and comparative evaluations. arXiv preprint arXiv:2104.10824, 2021
- ［ 8 ］ Liu WN, Zhang YY, Bian XQ, et al.: Study on detection rate of polyps and adenomas in artificial-intelligence-aided colonoscopy. Saudi J. Gastroenterol 26: 13-19, 2020
- ［ 9 ］ Lee S, Bae JS, Kim H, et al.: Liver lesion detection from weakly-labeled multi-phase CT volumes with a grouped Single Shot MultiBox Detector. In Proceedings of the 28th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Granada, 2018, LNCS Vol.11071, pp. 693-701
- ［10］ Das A, Acharya UR, Panda SS, et al.: Deep learning based liver cancer detection using watershed transform and Gaussian mixture model techniques. Cogn Syst Res 54: 165-175, 2019
- ［11］ 森 秀明，平井都始子：レジデント・臨床検査技師のための初めての超音波検査第2版．文光堂，2019，p. 79
- ［12］ Nishida N, Yamakawa M, Shiina T, et al.: Artificial intelligence (AI) models for the ultrasonographic diagnosis of liver tumors and comparison of diagnostic accuracies between AI and human experts. J Gastroenterol 57: 309-321, 2022
- ［13］ Nakashima T, Tsutsumi I, Takami H, et al.: A study on liver tumor detection from an ultrasound image using deep learning. In Proceedings the International Workshop on Advanced Image Technology （IWAIT） 2020, Yogyakarta, 2020, 115151v
- ［14］ Redmon J, Farhadi A: YOLOv3: An Incremental improvement. arXiv preprint arXiv:1804.02767, 2018
- ［15］ Itoh H, Misawa M, Mori Y, et al.: Positive- gradient weighted object activation mapping: Visual explanation of object detector towards precise colorectal-polyp localisation, International J Comput Assist Radiol Surg 17: 2051-2063, 2022
- ［16］ Lin Y, Maire M, Belongie S, et al.: Microsoft COCO: Common objects in context. In Proceedings of the 13th European Conference on Computer Vision (ECCV), Zurich, 2014, pp. 740-755
- ［17］ Deng J, Dong W, Socher R, et al.: Imagenet: A largescale hierarchical image database. In Proceedings of 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Miami, 2009, pp. 248-255
