Search for: All records

This working note documents the participation of CS_Morgan in the ImageCLEFmedical 2024 Caption subtasks, focusing on Caption Prediction and Concept Detection challenges. The primary objectives included training, validating, and testing multimodal Artificial Intelligence (AI) models intended to automate the process of generating captions and identifying multi-concepts of radiology images. The dataset used is a subset of the Radiology Objects in COntext version 2 (ROCOv2) dataset and contains image-caption pairs and corresponding Unified Medical Language System (UMLS) concepts. To address the caption prediction challenge, different variants of the Large Language and Vision Assistant (LLaVA) models were experimented with, tailoring them for the medical domain. Additionally, a lightweight Large Multimodal Model (LMM), and MoonDream2, a small Vision Language Model (VLM), were explored. The former is the instruct variant of the Image-aware Decoder Enhanced à la Flamingo with Interleaved Cross-attentionS (IDEFICS) 9B obtained through quantization. Besides LMMs, conventional encoder-decoder models like Vision Generative Pre-trained Transformer 2 (visionGPT2) and Convolutional Neural Network-Transformer (CNN-Transformer) architectures were considered. Consequently, this enabled 10 submissions for the caption prediction task, with the first submission of LLaVA 1.6 on the Mistral 7B weights securing the 2nd position among the participants. This model was adapted using 40.1M parameters and achieved the best performance on the test data across the performance metrics of BERTScore (0.628059), ROUGE (0.250801), BLEU-1 (0.209298), BLEURT (0.317385), METEOR (0.092682), CIDEr (0.245029), and RefCLIPScore (0.815534). For the concept detection task, our single submission based on the ConvMixer architecture—a hybrid approach leveraging CNN and Transformer advantages—ranked 9th with an F1-score of 0.107645. Overall, the evaluations on the test data for the caption prediction task submissions suggest that LMMs, quantized LMMs, and small VLMs, when adapted and selectively fine-tuned using fewer parameters, have ample potential for understanding medical concepts present in images. 

In the ever-changing realm of medical image processing, ImageCLEF brought a newdimension with the Identifying GAN Fingerprint task, catering to the advancement of visual media analysis. This year, the author presented the task of detecting training image fingerprints to control the quality of synthetic images for the second time (as task 1) and introduced the task of detecting generative model fingerprints for the first time (as task 2). Both tasks are aimed at discerning these fingerprints from images, on both real training images and the generative models. The dataset utilized encompassed 3D CT images of lung tuberculosis patients, with the development dataset featuring a mix of real and generated images, and the test dataset. Our team ’CSMorgan’ contributed several approaches, leveraging multiformer (combined feature extracted using BLIP2 and DINOv2) networks, additive and mode thresholding techniques, and late fusion methodologies, bolstered by morphological operations. In Task 1, our optimal performance was attained through a late fusion-based reranking strategy, achieving an F1 score of 0.51, while the additive average thresholding approach closely followed with a score of 0.504. In Task 2, our multiformer model garnered an impressive Adjusted Rand Index (ARI) score of 0.90, and a fine-tuned variant of the multiformer yielded a score of 0.8137. These outcomes underscore the efficacy of the multiformer-based approach in accurately discerning both real image and generative model fingerprints. 

The field of computer vision plays a key role in managing, processing, analyzing, and interpreting multimedia data in diverse applications. Visual interestingness in multimedia contents is crucial for many practical applications, such as search and recommendation. Determining the interestingness of a particular piece of media content and selecting the highest-value item in terms of content analysis, viewers’ perspective, content classification, and scoring media are sophisticated tasks to perform due to the heavily subjective nature. This work presents the approaches of the CS_Morgan team by participating in the media interestingness prediction task under ImageCLEFfusion 2023 benchmark evaluation. We experimented with two ensemble methods which contain a dense architecture and a gradient boosting scaled architecture. For the dense architecture, several hyperparameters tunings are performed and the output scores of all the inducers after the dense layers are combined using min-max rule. The gradient boost estimator provides an additive model in staged forward propagation, which allows an optimized loss function. For every step in the ensemble gradient boosting scaled (EGBS) architecture, a regression tree is fitted to the negative gradient of the loss function. We achieved the best accuracy with a MAP@10 score of 0.1287 by using the ensemble EGBS.