# Tackling the Incomplete Annotation Issue in Universal Lesion Detection Task By Exploratory Training

arXiv: https://arxiv.org/abs/2309.13306 · HTML: https://mlsanity.com/paper/2309.13306/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 30 references that are in the map.
- First posted on arXiv 23 Sep 2023.

## Builds on (most cited first, 23 of 23)

- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [A Closer Look at Memorization in Deep Networks](https://mlsanity.com/paper/1706.05394/) (2017)
- [Training Region-based Object Detectors with Online Hard Example Mining](https://mlsanity.com/paper/1604.03540/) (2016)
- [The Liver Tumor Segmentation Benchmark (LiTS)](https://mlsanity.com/paper/1901.04056/) (2019)
- [Unbiased Teacher for Semi-Supervised Object Detection](https://mlsanity.com/paper/2102.09480/) (2021)
- [SGD on Neural Networks Learns Functions of Increasing Complexity](https://mlsanity.com/paper/1905.11604/) (2019)
- [Rethinking Classification and Localization for Object Detection](https://mlsanity.com/paper/1904.06493/) (2019)
- [Learning from Multiple Datasets with Heterogeneous and Partial Labels for Universal Lesion Detection in CT](https://mlsanity.com/paper/2009.02577/) (2020)
- [MULAN: Multitask Universal Lesion Analysis Network for Joint Lesion Detection, Tagging, and Segmentation](https://mlsanity.com/paper/1908.04373/) (2019)
- [Improving RetinaNet for CT Lesion Detection with Dense Masks from Weak RECIST Labels](https://mlsanity.com/paper/1906.02283/) (2019)
- [Co-mining: Self-Supervised Learning for Sparsely Annotated Object Detection](https://mlsanity.com/paper/2012.01950/) (2020)
- [Solving Missing-Annotation Object Detection with Background Recalibration Loss](https://mlsanity.com/paper/2002.05274/) (2020)
- [3D Context Enhanced Region-based Convolutional Neural Network for End-to-End Lesion Detection](https://mlsanity.com/paper/1806.09648/) (2018)
- [Modeling Missing Annotations for Incremental Learning in Object Detection](https://mlsanity.com/paper/2204.08766/) (2022)
- [Soft Sampling for Robust Object Detection](https://mlsanity.com/paper/1806.06986/) (2018)
- [Object Detection as a Positive-Unlabeled Problem](https://mlsanity.com/paper/2002.04672/) (2020)
- [SATr: Slice Attention with Transformer for Universal Lesion Detection](https://mlsanity.com/paper/2203.07373/) (2022)
- [Asymmetric 3D Context Fusion for Universal Lesion Detection](https://mlsanity.com/paper/2109.08684/) (2021)
- [Improving Deep Lesion Detection Using 3D Contextual and Spatial Attention](https://mlsanity.com/paper/1907.04052/) (2019)
- [Lesion Harvester: Iteratively Mining Unlabeled Lesions and Hard-Negative Examples at Scale](https://mlsanity.com/paper/2001.07776/) (2020)
- [Revisiting 3D Context Modeling with Supervised Pre-training for Universal Lesion Detection in CT Slices](https://mlsanity.com/paper/2012.08770/) (2020)
- [Positive-unlabeled Learning for Cell Detection in Histopathology Images with Incomplete Annotations](https://mlsanity.com/paper/2106.15918/) (2021)
- [Decoupled Gradient Harmonized Detector for Partial Annotation: Application to Signet Ring Cell Detection](https://mlsanity.com/paper/2004.04455/) (2020)

## Built on by (most cited first, 1 of 1)

- [MalariAI: A Label-Resilient Decoupled Framework for Annotation-Agnostic Cell Segmentation and Explainable Stage Classification in Dense Malaria Blood Smears](https://mlsanity.com/paper/2607.00385/) (2026)

## Abstract (arXiv)

Universal lesion detection has great value for clinical practice as it aims to detect various types of lesions in multiple organs on medical images. Deep learning methods have shown promising results, but demanding large volumes of annotated data for training. However, annotating medical images is costly and requires specialized knowledge. The diverse forms and contrasts of objects in medical images make fully annotation even more challenging, resulting in incomplete annotations. Directly training ULD detectors on such datasets can yield suboptimal results. Pseudo-label-based methods examine the training data and mine unlabelled objects for retraining, which have shown to be effective to tackle this issue. Presently, top-performing methods rely on a dynamic label-mining mechanism, operating at the mini-batch level. However, the model's performance varies at different iterations, leading to inconsistencies in the quality of the mined labels and limits their performance enhancement. Inspired by the observation that deep models learn concepts with increasing complexity, we introduce an innovative exploratory training to assess the reliability of mined lesions over time. Specifically, we introduce a teacher-student detection model as basis, where the teacher's predictions are combined with incomplete annotations to train the student. Additionally, we design a prediction bank to record high-confidence predictions. Each sample is trained several times, allowing us to get a sequence of records for each sample. If a prediction consistently appears in the record sequence, it is likely to be a true object, otherwise it may just a noise. This serves as a crucial criterion for selecting reliable mined lesions for retraining. Our experimental results substantiate that the proposed framework surpasses state-of-the-art methods on two medical image datasets, demonstrating its superior performance.