While the Segment Anything Model (SAM) has demonstrated remarkable capabilities in general image segmentation, its direct application to medical images faces significant challenges, leading to domain-specific adaptations like MedSAM. In current medical literature, SAM-like models are often evaluated using “simulated” bounding box prompts derived from ground truth annotations, potentially overestimating their real-world performance. This paper addresses this underexplored problem by systematically evaluating the influence of bounding box quality on melanoma segmentation performance using the ISIC 2018 dataset (RGB skin image corpus). Experiments were conducted on both the foundational SAM and MedSAM, utilizing a structured protocol that includes ideal bounding boxes, baseline heuristic prompts (full image, most probable location, centered point for SAM), and controlled perturbations. These perturbations comprise five levels of expanding, shrinking, and shifting the ideal bounding box (random disturbance between 5% and 25%). Our results demonstrate a significant dependence of segmentation performance on bounding box quality for both models. Depending on the type of bounding box perturbation, the Dice score decreases from 84% under ideal conditions to 37% with inaccurate bounding boxes, corresponding to a reduction of more than 50%. Notably, shrinking the bounding box has a more detrimental effect than expanding it, with accuracy dropping substantially (e.g., MedSAM's Dice score decreases from 91.5 to 60.9 under a 25% shrinkage). Moreover, medical pre-training significantly improves robustness, as MedSAM outperforms SAM by approximately 7 Dice points under ideal conditions. These findings highlight that although SAM-like models offer promising capabilities, their reliance on high-quality bounding box prompts currently limits their applicability for fully unsupervised segmentation in medical image processing pipelines. This underscores the need for either human intervention or more advanced automated prompting strategies in clinical applications.

Segmentation of medical images using learning-based systems remains a challenge in medical computer vision, as training a segmentation model requires exhaustively annotated medical images by experts, which are difficult and expensive to obtain. In the context of melanoma segmentation, we explored active learning methods to combine human annotation for the most uncertain pixels with model predictions for the others. With only around$30 \%$of the images annotated, we achieved performance similar to that obtained with a fully annotated dataset. We also demonstrated that, after a few iterations, experts can focus on annotating only the most uncertain areas of the images, relying on the model for the rest. These approaches pave the way for accelerating the annotation of unlabeled medical datasets and optimizing the use of medical expertise in deep learning projects.

Segmentation of medical images using learning based systems remains a challenge in medical computer vision: training a segmentation model requires medical images exhaustively annotated by experts that are difficult and expensive to obtain. We propose to explore the usage of partially annotated images, i.e., all images are annotated but not all regions of a given class are annotated. In this paper, we propose several approaches and we experiment them on the segmentation of intra-oral images. First, we propose to modify the loss function to consider only the annotated areas, and second to integrate annotation from non-expert, as well as the combination of these methods. The experiments we conducted showed an improvement up to 33% on the segmentation performance. This approach allows to obtain better quality annotation masks than the initial human annotation using only partially annotated areas or non-expert annotations. In the future, these approaches can be extended by combination with active learning methods.

Over the past years, deep learning methods allowed for new state-of-the-art results in ad-hoc information retrieval. However such methods usually require large amounts of annotated data to be effective. Since most standard ad-hoc information retrieval datasets publicly available for academic research (e.g. Robust04, ClueWeb09) have at most 250 annotated queries, the recent deep learning models for information retrieval perform poorly on these datasets. These models (e.g. DUET, Conv-KNRM) are trained and evaluated on data collected from commercial search engines not publicly available for academic research which is a problem for reproducibility and the advancement of research. In this paper, we propose WIKIR: an open-source toolkit to automatically build large-scale English information retrieval datasets based on Wikipedia. WIKIR is publicly available on GitHub. We also provide wikIR59k: a large-scale publicly available dataset that contains 59,252 queries and 2,617,003 (query, relevant documents) pairs.

Document indexing is a key component for efficient information retrieval (IR). After preprocessing steps such as stemming and stop-word removal, document indexes usually store term-frequencies (tf). Along with tf (that only reflects the importance of a term in a document), traditional IR models use term discrimination values (TDVs) such as inverse document frequency (idf) to favor discriminative terms during retrieval. In this work, we propose to learn TDVs for document indexing with shallow neural networks that approximate traditional IR ranking functions such as TF-IDF and BM25. Our proposal outperforms, both in terms of nDCG and recall, traditional approaches, even with few positively labelled query-document pairs as learning data. Our learned TDVs, when used to filter out terms of the vocabulary that have zero discrimination value, allow to both significantly lower the memory footprint of the inverted index and speed up the retrieval process (BM25 is up to 3~times faster), without degrading retrieval quality.

Logic-based information retrieval (IR) models represent the retrieval decision as an implication d→q between a document d and a query q, where d and q are logical sentences. However, d→q is generally a binary decision, thus we need a measurement to estimate the degree to which d implies q, denoted U(d→q)⁠. Most of the existing logic-based IR models either do not precisely define the implication d→q or use non-classical definitions. Some models also define the uncertainty U in informal ways. More importantly, they use two non-related frameworks to define d→q and its uncertainty U, even though the two notions are intrinsically related. The goal of this study is to propose a new logic-based IR model, which overcome these shortcomings. To this end, we first propose to replace the implication d→q by the validity of material implication⊨d⊃q⁠. Second, we redefine and adapt the mathematical relationship between logics, lattices and probability. Our new IR model presents a possible formalism for van Rijsbergen's intuition about replacing U(d→q) by P(q∣d)⁠.

Logic-based Information Retrieval (IR) models represent the retrieval decision as a logical implication d->q between a document d and a query q, where d and q are logical sentences. However, d->q is a binary decision, we thus need a measure to estimate the degree to which d implies q, denoted P(d->q). In this study, we revisit the Van Rijsbergen's assumptions about: 1- the logical implication ->' is not the material one, and 2- P(d->q) could be estimated by the conditional probability P(q|d). More precisely, we claim that the material implication is an appropriate implication for IR, and also we mathematically prove that replacing P(d->q) by P(q|d) is a correct choice. In order to prove the Van Rijsbergen's assumption, we use the Propositional Logic and the Lattice theory. We also exploit the notion of degree of implication that is proposed by Knuth.

The quality of semantic tuples (semantic triples forming subject-predicate-object) has significant impact in most text mining and knowledge discovery applications. The practical success and usability of these applications momentously depends on the quality of the extracted semantic triples. Most biomedical semantic resources have been developed for different contexts focusing on the structural representation but with less attention on the acceptability and naturalness of the individual semantic triples. In this article, we presented an integrated approach for enhancing the quality of semantic tuples in the UMLS knowledge sources. The approach is based on the integration of three existing auditing techniques: avoiding redundant classifications of semantic concepts, reducing hierarchical and associative relationship inconsistencies. We evaluated the approach based on the number of identified wrongly assigned concepts and inconsistent relationships obtained. The quality of each semantic triple is evaluated based on the acceptability and naturalness of the semantic tuples. The evaluation shows promising results. In the evaluation, we have extracted 10,082 semantic triples randomly from UMLS and obtained 5646 taxonomically and 4436 non-taxonomically related semantic triples. 826 concepts are found redundantly classified and 352 are found hierarchically inconsistent. In non-taxonomic semantic triples, out of 4436, 726 are found to be inconsistent. The quality (acceptability and naturalness) of each semantic triples of the first 100 are also evaluated using domain experts. The Cohen's kappa coefficient is used to measure the degree of agreement between the annotators and the result is promising (0.8).

In this paper, we propose an algorithm for learning a general class of similarity measures for kNN classification. This class encompasses, among others, the standard cosine measure, as well as the Dice and Jaccard coefficients. The algorithm we propose is an extension of the voted perceptron algorithm and allows one to learn different types of similarity functions (either based on diagonal, symmetric or asymmetric similarity matrices). The results we obtained show that learning similarity measures yields significant improvements on several collections, for two prediction rules: the standard kNN rule, which was our primary goal, and a symmetric version of it.

With the current trend, we can anticipate that future mobile phones will have ever-increasing computational power and be able to embed several captors/effectors including cameras, GPS, orientation sensors, tactile surfaces and vibro-tactile display. Such powerful mobile platforms enable us to deploy mixed reality systems. Many studies on mobile mixed reality focus on games. In this paper, we describe the deployment and a user study of a mixed reality location-based mobile treasure hunt, Snap2Play[1], using technologies such as place recognition, accelerometers and GPS tracking for enhancing the interaction with the game and therefore the game playability. The game that we deployed and tested is running on an off-the-shelf camera phone.

The task of ad hoc photographic image retrieval in ImageCLEF 2007 international benchmark is to retrieve relevant images in the database to the user query formulated as keywords and image examples. This paper presents rich representation and indexing technologies exploited in our system that participated in ImageCLEF 2007. It uses diverse visual content representation, text representation, pseudo-relevance feedback and fusion, which make our system, with mean average precision 0.2833, in the 4th place among 457 automatic runs submitted from 20 participants to photographic ImageCLEF 2007 and in the 2nd place in terms of participants. Our systematic analysis in the paper demonstrates that 1) combing diverse low-level visual features and ranking technologies significantly improves the content-based image retrieval (CBIR) system; 2) cross-modality pseudo-relevance feedback improves the system performance; and 3) fusion of CBIR and TBIR outperforms individual modality based system.

Conceptual Indexing is a way to produce only one index for many multilingual documents. Inter-Media conceptual indexing promotes the use of common concepts between two media in order to use a single index for several media. In this paper we explore such an advance indexing point of view. We show the benefit of an automatic conceptual indexing for texts and its extension for text and image documents. Tests are conducted on the multilingual image and text medical document corpus of the CLEF initiative, where we obtain best results on text in 2005 and 2006, and show promising results on images, and best results for the combination of image and text.

This paper studies the effect of Latent Semantic Analysis (LSA) on two different tasks: multimedia document retrieval (MDR) and automatic image annotation (AIA). The contributions of this paper are twofold. First, to the best of our knowledge, this work is the first study of the influence of LSA on the retrieval of a significant number of multimedia documents (i.e. collection of 20000 tourist images). Second, it shows how different image representations (region-based and keypoint-based) can be combined by LSA to improve automatic image annotation. The document collections used for these experiments are the Corel photo collection and ImageCLEF 2006 collection.

In this paper, we present a knowledge-assisted approach to index and retrieve large volume of medical images. Both images and associated texts are indexed using medical concepts from the Unified Medical Language System (UMLS) meta-thesaurus. We propose a structured learning framework for modular acquisition of medical semantics from images with complementary global and local image indexing schemes. Two fusion approaches are also developed to improve text retrieval using the UMLS-based image indexing: a simple post-query fusion and a visual modality filtering to remove visually aberrant images according to the query modality concepts. On the ImageCLEFmed 2005 database, our framework outperformed our previous result which ranked top in the ImageCLEFmed 2005 Medical Image Retrieval task benchmark.

Camera phones present new opportunities and challenges for mobile information association and retrieval. The visual input in the real environment is a new and rich interaction modality between a mobile user and vast information base connected to a user's device via rapidly advancing communication infrastructure. We have developed a system for tourist information access to provide scene description based on an image taken of the scene. In this paper, we describe the working system, the STOIC 101 database, and a new pattern discovery algorithm to learn image patches that are recurrent within a scene class and discriminative across others. We report preliminary scene recognition results on 90 scenes, trained on 5 images per scene, with an accuracy of 92% and 88% on a test set of 110 images, with and without location priming.

In this paper, we revisit the dependence language modelfor information retrieval proposed in [1], and show that thismodel is deficient from a theoretical point of view. We thenpropose a new model, well founded theoretically, for integratingdependencies between terms in the language model.This new model is simpler, yet more general, than the oneproposed in [1], and yields similar results in our experiments,on both syntactic and semantic dependencies.

Voluminous medical images are generated daily. They are critical assets for medical diagnosis, research, and teaching. To facilitate automatic indexing and retrieval of large medical-image databases, both images and associated texts are indexed using medical concepts from the Unified Medical Language System (UMLS) meta-thesaurus. We propose a structured learning framework based on support vector machines to facilitate modular design and learning of medical semantics from images. We present two complementary visual indexing approaches within this framework: a global indexing to access image modality and a local indexing to access semantic local features. Two fusion approaches are developed to improve textual retrieval using the UMLS-based image indexing. First, a simple fusion of the textual and visual retrieval approaches is proposed, improving significantly the retrieval results of both text and image retrieval. Second, a visual modality filtering is designed to remove visually aberrant images according to the query modality concept(s). Using the ImageCLEFmed database, we demonstrate the effectiveness of our framework which is superior when compared with the automatic runs evaluated in 2005 on the same medical-image retrieval task.

Weblog (usually shortened as blog) has gained its popularity lately. There are about 70,000 new blogs a day and about 29,100 blog updates an hour. As an emerging blogging phenomenon, with the proliferation of camera phones, mobile bloggers can write their blogs almost instantaneously. But how much further can current mobile blogging tools enhance the experience? In this paper, we propose a Mobilog framework to automate context-relevant annotation and synthesise personalised content for mobile blogging. In particular, we describe a system implementation of the framework, Travelog, adapted for tourism applications. Finally, we discuss the challenges and future possibilities for mobile blogging

In order to solve medical multimodal queries, we propose to split the queries in different dimensions using ontology. We extract both textual and visual terms depending on the ontology dimension they belong to. Based on these terms, we build different sub queries each corresponds to one query dimension. Then we use Boolean expressions on these sub queries to filter the entire document collection. The filtered document set is ranked using the techniques in vector space model. We also combine the ranked lists generated using both text and image indexes to further improve the retrieval performance. We have achieved the best overall performance for the medical image retrieval task in CLEF 2005. These experimental results show that while most queries are better handled by the text query processing as most semantic information are contained in the medical text cases, both textual and visual ontology dimensions are complementary in improving the results during media fusion

We propose a novel framework for content-based image retrieval with multiple parallel retrieval engines (MultiPRE) to achieve higher retrieval performance. Visual features, including both low-level features, such as color, texture and region features, and middle-level structure features, such as blob representation of objects are used to capture geometrical and statistical characteristics of images. Both clustering analysis and discrimination analysis are used as similarity measures in multiple retrieval engines, which are based on~principal component analysis (PCA) and support vector machines (SVM), respectively. Finally outputs of these engines are fused to determine ranking lists of retrieved images for given retrieval topics. The proposed framework has been evaluated based on the 26 image query topics over the CasImage database~with over 9000 medical images~used in ImageCLEF 2004, an international research effort for content-based image retrieval performance benchmark. Experiments show that the proposed framework achieved significantly better performance in terms of both the mean and the variance of average precision than the best run reported in ImageCLEF2004.

In this paper we investigate some aspects of the logical approach for retrieval models. We develop here the idea that the intrinsic expressive power of logic has already brought its benefits in information retrieval through the design of the first logical retrieval model which has proved more general, or powerful, than any existing model. This benefit lies in the expressive power, or generality of logic on one side, and in its very close fitness with the fundamentals of information retrieval. We mainly develop this second point in giving an informal interpretation of the semantics of the logical model as it was first introduced by C. J. Van Rijsbergen in 1986. This discussion is also for us an occasion to investigate in more details some less known aspects of information retrieval, as for example the impact of new applications which already induce a complete revision of the notion of document. Then the logical model is presented as the one the expressive power of which may provide a much needed common framework for a coherent integration of recent approaches in information retrieval involving the use of Natural Language Processing techniques (NLP) or Artificial Intelligence (AI) techniques. This idea is further developed in showing that the logical approach may also help in giving new insights into classical retrieval models and even help in improving them in a coherent way. An example of this later idea is developed with the Boolean model.