Pasar al contenido principal

RECUPERACION DE IMAGENES BASADA EN EL CONTENIDO PARA EL DIAGNOSTICO DE TUMORES CUTANEOS PRIMARIOS Y SECUNDARIOS

PID2022-140189OB-C21

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2022
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITÈCNICA DE VALÈNCIA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 15
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

MIL-Adapter: Coupling multiple instance learning and vision-language adapters for few-shot slide-level classification

  • Meseguer-Esbrí, Pablo|||0000-0001-7821-6168
  • del Amor, Rocío
  • Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
[EN] Contrastive language-image pretraining has greatly enhanced visual representation learning and enabled zero-shot classification. Vision-language language models (VLM) have succeeded in few-shot learning by leveraging adaptation modules fine-tuned for specific downstream tasks. In computational pathology (CPath), accurate whole-slide image (WSI) prediction is crucial for aiding in cancer diagnosis, and multiple instance learning (MIL) remains essential for managing the gigapixel scale of WSIs. In the intersection between CPath and VLMs, the literature still lacks specific adapters that handle the particular complexity of the slides. To solve this gap, we introduce MIL-Adapter, a novel approach designed to obtain consistent slide-level classification under few-shot learning scenarios. In particular, our framework is the first to combine trainable MIL aggregation functions and lightweight visual-language adapters to improve the performance of histopathological VLMs. MIL-Adapter relies on textual ensemble learning to construct discriminative zero-shot prototypes. It is serves as a solid starting point, surpassing MIL models with randomly initialized classifiers in data-constrained settings. With our experimentation, we demonstrate the value of textual ensemble learning and the robust predictive performance of MIL-Adapter through diverse datasets and configurations of few-shot scenarios, while providing crucial insights on model interpretability. The code is publicly accessible in https://github.com/cvblab/MIL-Adapter., This work has received funding from the Spanish Ministry of Economy and Competitiveness through the projects PID2022-140189OB-C21 (ASSIST) and CIPROM/2022/20 (COMTACTS2) .
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A fusocelular skin dataset with whole slide images for deep learning models

  • del Amor, Rocío
  • López-Pérez, Miguel|||0000-0003-2965-0624
  • Meseguer-Esbrí, Pablo|||0000-0001-7821-6168
  • Morales, Sandra|||0000-0003-0763-1545
  • Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
  • Terradez, Liria
  • Aneiros-Fernandez, Jose
  • Mateos, Javier
  • Molina, Rafael
[EN] Cutaneous spindle cell (CSC) lesions encompass a spectrum from benign to malignant neoplasms, often posing significant diagnostic challenges. Computer-aided diagnosis systems offer a promising solution to make pathologists' decisions objective and faster. These systems usually require large-scale datasets with curated labels for effective training; however, manual annotation is time-consuming and expensive. To overcome this challenge, crowdsourcing has emerged as a popular and valuable strategy to scale up the labeling process by distributing the effort among different non-expert annotators. This work introduces AI4SkIN, the first public dataset Whole Slide Images (WSIs) for CSC neoplasms, annotated using an innovative crowdsourcing protocol. AI4SkIN dataset contains 641 Hematoxylin and Eosin stained WSIs with multiclass labels from both expert and trainee pathologists. The dataset improves CSC neoplasm diagnosis using advanced machine learning and crowdsourcing based on Gaussian Processes, showing that models trained on non-expert labels perform comparably to those using expert labels. In conclusion, we illustrate that AI4SkIN provides a good resource for developing and validating methods for multiclass CSC neoplasm classification., This work has received funding from the Spanish Ministry of Economy and Competitiveness through project PID2019-105142RB-C21 and PID2019-105142RB-C22 (AI4SkIN) and PID2022-140189OB-C21 and PID2022-140189OB-C22 (ASSIST), and Generalitat Valenciana through project COM-TACTS2 (CIPROM/2022/20). The work of Rocio del Amor was supported by the Spanish Ministry of Universities (FPU20/05263). The work of Pablo Meseguer was funded by valgrAI - Valencian Graduate School and Research Network of Artificial Intelligence. The work of Sandra Morales was co-funded by the Universitat Politecnica de Valencia through the program PAID-10-20. The work of M. Lopez-Perez was supported by the grant JDC2022-048318-I funded by MICIU/AEI/10.13039/501100011033 and the European Union NextGenerationEU/PRTR.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images

  • Meseguer-Esbrí, Pablo|||0000-0001-7821-6168
  • del Amor, Rocío
  • Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
[EN] Artificial intelligence (AI) agents encounter the problem of catastrophic forgetting when they are trained in sequentially with new data batches. This issue poses a barrier to the implementation of AI -based models in tasks that involve ongoing evolution, such as cancer prediction. Moreover, whole slide images (WSI) play a crucial role in cancer management, and their automated analysis has become increasingly popular in assisting pathologists during the diagnosis process. Incremental learning (IL) techniques aim to develop algorithms capable of retaining previously acquired information while also acquiring new insights to predict future data. Deep IL techniques need to address the challenges posed by the gigapixel scale of WSIs, which often necessitates the use of multiple instance learning (MIL) frameworks. In this paper, we introduce an IL algorithm tailored for analyzing WSIs within a MIL paradigm. The proposed M ultiple I nstance C lass- I ncremental L earning ( MICIL ) algorithm combines MIL with class-IL for the first time, allowing for the incremental prediction of multiple skin cancer subtypes from WSIs within a class-IL scenario. Our framework incorporates knowledge distillation and data rehearsal, along with a novel embedding-level distillation, aiming to preserve the latent space at the aggregated WSI level. Results demonstrate the algorithm's effectiveness in addressing the challenge of balancing IL-specific metrics, such as intransigence and forgetting, and solving the plasticity-stability dilemma., This work has received funding from Horizon 2020, the European Union's Framework Programme for Research and Innovation, under grant agreement No. 860627 (CLARIFY), the Spanish Ministry of Economy and Competitiveness through projects PID2019-105142RB-C21(AI4SKIN) and PID2022-140189OB-C21 (ASSIST), and GVA through the project INNEST/2021/321 (SAMUEL). The work of Rocio del Amorand Pablo Meseguer has been supported by the Spanish Ministry of Universities under an FPU Grant (FPU20/05263) and valgrAI - Valencian Graduate School and Research Network of Artificial Intelligence, respectively.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Fuzzy Inference Systems to Fine-Tune a Local Eigenvector Image Smoothing Method

  • Almutairi, Khleef
  • Latorre-Carmona, Pedro
  • Morillas, Samuel|||0000-0001-9262-6139
[EN] Image denoising is a fundamental research topic in colour image processing, analysis, and transmission. Noise is an inevitable byproduct of image acquisition and transmission, and its nature is intimately linked to the underlying processes that produce it. Gaussian noise is a particularly prevalent type of noise that necessitates effective removal while ensuring the preservation of the original image's quality. This paper presents a colour image denoising framework that integrates fuzzy inference systems (FISs) with eigenvector analysis. This framework employs eigenvector analysis to extract relevant information from local image neighbourhoods. This information is subsequently fed into the FIS system which dynamically adjusts the intensity of the denoising process based on local characteristics. This approach recognizes that homogeneous areas may require less aggressive smoothing than detailed image regions. Images are converted from the RGB domain to an eigenvector-based space for smoothing and then converted back to the RGB domain. The effectiveness of the proposed methods is established through the application of various image quality metrics and visual comparisons against established state-of-the-art techniques., This research was funded by Generalitat Valenciana under grant CIAICO/2022-051 IMaLeVICS and Spanish Ministry of Science under grant PID2022-140189OB-C21.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Generalizing a Construction of Non-Strong Fuzzy Metrics from Metrics and Studying Their Induced Topology

  • Grigorenko, Olga
  • Talia, Simona
  • Miñana, Juan-José|||0000-0001-9835-0700
[EN] The problem of obtaining new examples of fuzzy metrics is of interest, as this type of fuzzy measurement has been proven to be useful in engineering applications. In this context, different works have addressed the problem of deriving fuzzy metrics from classical ones. This paper is devoted to generalizing a construction of non-strong fuzzy metrics from metrics already provided in the literature, both for continuous Archimedean t-norms and for the minimum t-norm. Moreover, we explore the conditions under which one adapts this generalized method to obtain fuzzy metrics in the sense of George and Veeramani. In addition, we investigate the connection between the topology associated with the fuzzy metric constructed via these procedures and that determined by the metric. Several examples are provided to support and illustrate our findings., Financiado con Ayuda a Primeros Proyectos de Investigacion (PAID-06-24), Vicerrectorado de Investigacion de la Universitat Politecnica de Valencia (UPV). This research is part of projects PID2022-139248NB-I00 and PID2022-140189OB-C21 funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, and CIAICO/2022/051 funded by Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

The puzzling Spitz tumours: is artificial intelligence the key to their understanding?

  • Launet, Laetitia Mariana
  • Mosquera, Andrés
  • Monteagudo, Carlos
  • Colomer, Adrián|||0000-0002-7616-6029
  • Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
[EN] Since their first description in 1948, Spitz tumours remain one of the most challenging diagnostic entities in dermatopathology due to their complex histological features and ambiguous clinical behaviour. In recent years, artificial intelligence (AI) solutions have demonstrated significant potential across a wide range of medical applications, including computational pathology, for decision-making in diagnosis, along with promising advances in prognosis and tumour classification., This work received funding from projects CIPROM/2022/20 (COMTACTS2), PID2022-140189OB-C21(ASSIST) and PI20/00094, Instituto de Salud CarlosIII, FEDER European Funds.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Indistinguishability operators and generalized distances: the transformation problem revisited

  • González-Hedström, J. D. D.
  • Valero, O.
  • Miñana, Juan-José|||0000-0001-9835-0700
[EN] In this paper we characterize those functions that induce a fuzzy preorder from a quasi-pseudo-metric even when the considered t-norm is not continuous. On the one hand, we prove that they must be decreasing and fulfill a special property of dominance with respect to the ordinary addition and the t-norm T under consideration. On the other hand, we have shown that they must transform asymmetric triangular triplets into asymmetric T-triangular triplets. Moreover, we also study the case in which fuzzy preorders are exactly indistinguishability operators. Concretely, we show that the monotonicity of the function and the previously mentioned dominance are sufficient but not necessary conditions. In addition, we prove that such functions must transform triangular triplets into T-triangular triplets. The developed theory is illustrated by means of appropriate examples. Furthermore, we prove that the well-known technique based on the use of the pseudo-inverses of the additive generators of continuous and Archimedean t-norms is recovered as a particular case of the new one presented here., This research is part of projects PID2022-139248NB-I00 and PID2022-140189OB-C21 funded by MICIU/AEI/10.13039/- 501100011033 and ERDF/EU, CIAICO/2022/051 funded by Generalitat Valenciana. The first and third authors thank the support from Programa de Foment de la Recerca i la Innovacio de la Universitat de les Il les Balears 2024-2026 and ITS2023-086-Programa de Foment a la recerca. Financiado con Ayuda a Primeros Proyectos de Investigacion (PAID-06-24), Vicerrectorado de Investigacion de la Universitat Politecnica de Valencia (UPV).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A comparative analysis of machine learning methods for display characterization

  • Almutairi, Khleef
  • Latorre-Carmona, Pedro
  • Dansoko, Makan
  • Gacto, María José
  • Morillas, Samuel|||0000-0001-9262-6139
[EN] This paper explores the application of various machine learning methods for the characterization of LCD, OLED, and QLED display technologies, with the goal of achieving accurate color reproduction. The models are developed using input data (device-dependent RGB values) and output data (device-independent XYZ coordinates) collected from three different displays. Training and test datasets are built using measurements obtained directly from the displays, alongside corresponding color coordinates measured with a high-precision colorimeter. A key focus of this research is the use of fuzzy inference systems to build interpretable models. These models not only demonstrate strong performance in color reproduction but also provide valuable physical insights into the relationships between the RGB inputs and the resulting XYZ outputs. This interpretability allows for a deeper understanding of the display behavior. Furthermore, we compare the performance of fuzzy models with other widely used machine learning approaches, including neural networks and decision trees. By evaluating the strengths and weaknesses of each method, we aim to identify the most effective approach for display characterization. The effectiveness of each method is assessed based on its ability to accurately reproduce and display colors, measured using the ΔE₀₀ visual error metric. Our findings indicate that the Fuzzy Modeling and Identification (FMID) method is particularly effective, achieving an optimal balance between high accuracy and interpretability. Its competitive performance across all display types, combined with its ability to provide explanatory insights, suggests potential future applications in display calibration and optimization strategies. This study also investigates whether machine learning methods can offer advantages over traditional physical models, particularly in scenarios with limited data, and highlights the interpretability benefits of fuzzy inference systems in the context of display characterization., This work was supported by Generalitat Valenciana under grant IMaLeVICS CIAICO-2022-051 and Spanish Agencia Estatal de Investigacion under grant PID2022-140189OB-C21, PID2023-152301OB-I00. We would like to express our sincere gratitude to Eduardo Gomez Fernandez for his help in the experimental part of this study.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A new class of contractions on graphical fuzzy metric spaces and fixed point theorems

  • Shukla, Satish
  • Rai, Shweta
  • Miñana, Juan-José|||0000-0001-9835-0700
[EN] The aim of this paper is to introduce a new class of mappings called (H, H0 )-fuzzy graphical ¿-contractions in graphical fuzzy metric spaces. Some fixed point results for the mappings of this class are proved. This class of mappings and the fixed point results of this paper generalize and extend several fuzzy analogues of Banach contraction principle in graphical fuzzy metric spaces. Some examples and several consequences of the new concepts and results are given., The first author is grateful to Professor M.K. Dube for his regular encouragement and motivation for research. The first author is also thankful to the Science and Engineering Research Board (SERB), TAR/2022/000131, New Delhi, India, for their support. This research is part of projects PID2022-139248NBI00 and PID2022-140189OB-C21 funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, CIAICO/2022/051 funded by Generalitat Valenciana. Financiado con Ayuda a Primeros Proyectos de Investigacion (PAID-06-24), Vicerrectorado de Investigaci¿on de la Universitat Politecnica de Valencia (UPV).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Vector-Valued Fuzzy Metric Spaces and Fixed Point Theorems

  • Shukla, Satish
  • Dubey, Nikita
  • Miñana, Juan-José|||0000-0001-9835-0700
[EN] The purpose of this paper is to generalize the concept of classical fuzzy set to vector-valued fuzzy set which can attend values not only in the real interval [0, 1] , but in an ordered interval of a Banach algebra as well. This notion allows us to introduce the concept of vector-valued fuzzy metric space which generalizes, extends and unifies the notion of classical fuzzy metric space and complex-valued fuzzy metric space and permits us to consider the fuzzy sets and metrics in a larger domain. Some topological properties of such spaces are discussed and some fixed point results in this new setting are proved. Multifarious examples are presented which clarify and justify our claims and results., This research was funded by Science and Engineering Research Board (SERB) grant number TAR/2022/000131 and MCIN/AEI/10.13039/501100011033 y FEDER Una manera de hacer Europa under grants PID2022-140189OB-C21 and PID2022-139248NB-I00 and Generalitat Valenciana under grant CIAICO/2022/051 and European Union s Horizon 2020 research and innovation programme under grant agreements No. 871260.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

CLAV: Clustering Latent Vector Aggregation for Whole Slide Image Retrieval Leveraging Foundation Models

  • Golfe-San Martín, Alejandro|||0000-0001-5095-9904
  • Meseguer-Esbrí, Pablo|||0000-0001-7821-6168
  • Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
  • Colomer, Adrián|||0000-0002-7616-6029
[EN] Content-Based Image Retrieval (CBIR) is crucial in cancer diagnosis, assisting pathologists by providing similar image data from previous records for analysis, especially when there is uncertainty in diagnosing a case. This process supports decision-making by providing valuable reference points to guide the diagnostic process. Foundation models have become increasingly important in the medical field due to their ability to generalize across various tasks and datasets, offering valuable support to pathologists by enhancing the accuracy and efficiency of diagnostic processes. In this article, a foundation model pre-trained on histopathology data is leveraged as a feature extractor without the need for task-specific training, in contrast to existing models that require extensive training to learn significant data representations. The proposed method, Clustering Latent Vector Aggregation (CLAV), condenses the significant feature vectors into a unique representative vector for the Whole Slide Image (WSI). Using a unique feature vector offers the advantage of reducing the size of the memory bank, thereby making the process of querying and retrieving similar WSIs more efficient. The experimental results presented in this study demonstrate that the proposed method enhances performance in CBIR tasks. This article highlights the potential of foundation models to achieve superior retrieval metrics compared to state-of-the-art methods specifically trained for CBIR., This work has received funding from Spanish Ministry of Economy and Competitiveness through the project PID2022-140189OB-C21 (ASSIST).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

On the Axiomatic of GV-Fuzzy Metric Spaces and Its Completion

  • Gregori Gregori, Valentín|||0000-0002-5983-6182
  • Miñana, Juan-José|||0000-0001-9835-0700
  • Roig, Bernardino|||0000-0002-9599-572X
  • Sapena Piera, Almanzor|||0000-0001-8473-6063
[EN] The concept of fuzzy metric space introduced by Kramosil and Michalek was later slightly modified by George and Veeramani who imposed three additional restrictions on it. A significant difference between these two concepts of fuzzy metrics is that fuzzy metric spaces in the sense of George and Veeramani do not admit completion, in general. This paper is devoted to go into detail on completable fuzzy metric spaces by means of the study of the impact on the completion of each one of the restrictions imposed by George and Veeramani in their definition of fuzzy metric. In this direction, we characterize those completable fuzzy metric spaces, in which just one of the three restrictions imposed by George and Veeramani is required. Various examples illustrate and justify the main results., Financiado con Ayuda a Primeros Proyectos de Investigacion (PAID-06-24), Vicerrectorado de Investigacion de la Universitat Politecnica de Valencia (UPV). This research is part of projects PID2022-139248NB-I00 and PID2022-140189OB-C21 funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, CIAICO/2022/051 funded by Generalitat Valenciana and from project BUGWRIGHT2. This last project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreements No. 871260.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

On Completeness and Fixed Point Theorems in Fuzzy Metric Spaces

  • Gregori Gregori, Valentín|||0000-0002-5983-6182
  • Miñana, Juan-José|||0000-0001-9835-0700
  • Roig, Bernardino|||0000-0002-9599-572X
  • Sapena Piera, Almanzor|||0000-0001-8473-6063
[EN] This paper is devoted to showing the relevance of the notion of completeness used to establish a fixed point theorem in fuzzy metric spaces introduced by Kramosil and Michalek. Specifically, we show that demanding a stronger notion of completeness, called p-completeness, it is possible to relax some extra conditions on the space to obtain a fixed point theorem in this framework. To this end, we focus on a fixed point result, proved by Mihet for complete non-Archimedean fuzzy metric spaces (Theorem 1). So, we define a weaker concept than the non-Archimedean fuzzy metric, called t-strong, and we establish an alternative version of Mihe¿t¿s theorem for p-complete t-strong fuzzy metrics (Theorem 2). In addition, an example of t-strong fuzzy metric spaces that are not non-Archimedean is provided., This research was funded by Proyecto PID2022-139248NB-I00, Proyecto PID2022-140189OBC21 and Proyecto PID2021-126645OB-I00 financiados por MCIN/AEI/10.13039/501100011033 y FEDER Una manera de hacer Europa ; Proyecto CIAICO/2022/025 financiado por la Generalitat Valenciana; and from project BUGWRIGHT2. This last project has received funding from the European Union s Horizon 2020 research and innovation programme under grant agreements No. 871260. This publication reflects only the authors views and the European Union is not liable for any use that may be made of the information contained therein.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Enhancing Image Retrieval Performance With Generative Models in Siamese Networks

  • Golfe-San Martín, Alejandro|||0000-0001-5095-9904
  • Colomer, Adrián|||0000-0002-7616-6029
  • Prades Nebot, José|||0000-0003-4489-3704
  • Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
[EN] Prostate cancer is a critical healthcare challenge globally and is one of the most prevalent types of cancer in men. Early and accurate diagnosis is essential for effective treatment and improved patient outcomes. In the existing literature, computer-aided diagnosis (CAD) solutions have been developed to assist pathologists in various tasks, including classification, diagnosis, and prostate cancer grading. Content-based image retrieval (CBIR) techniques provide valuable approaches to enhance these computer-aided solutions. This study evaluates how generative deep learning models can improve the quality of retrievals within a CBIR system. Specifically, we propose applying a Siamese Network approach, which enables us to learn how to encode image patches into latent representations for retrieval purposes. We used the ProGleason-GAN framework trained on the SiCAPv2 dataset to create similar pairs of input patches. Our observations indicate that introducing synthetic patches leads to notable improvements in the evaluated metrics, underscoring the utility of generative models within CBIR tasks. Furthermore, this work is the first in the literature where latent representations optimized for CBIR are used to train an attention mechanism for performing Gleason Scoring of a WSI., The work of Alejandro Golfe was supported by the Valencian Graduate School and Research Network for Artificial Intelligence (valgrAI) . The work of Adrian Colomer was supported by "Ayuda aPrimeros Proyectos de Investigacion (PAID-06-22), Vicerrectorado deInvestigacion de la Universitat Politecnica de Valencia (UPV). This work was supported by the Spanish Ministry of Economy and Competitiveness through the Project PID2022-140189OB-C21 (ASSIST).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A new iterative inverse display model

  • Pérez Peñalver, María José|||0000-0003-0403-1405
  • Jordan-Lluch, Cristina|||0000-0001-9608-2984
  • Sanabria-Codesal, Esther|||0000-0002-4523-1991
  • Morillas, Samuel|||0000-0001-9262-6139
  • Lee, S. -W.
[EN] In this paper, we propose anew inverse model for display characterization based on the direct model developed in Kim and Lee (2015). We use an iterative method to compute what inputs are able to produce a desired color expressed in device independent color coordinates. Whereas iterative approaches have been used in the past for this task, the main novelty in our proposal is the use of specific heuristics based on the former display model and color science principles to achieve an efficient and accurate convergence. On the one hand, to set the initial point of the iterative process, we use orthogonal projections of the desired color chromaticity, xy, onto the display's chromaticity triangle to find the initial ratio the RGB coordinates need to have. Subsequently, we use a factor product, preserving RGB proportions, to initially approximate the desired color's luminance. This factor is obtained through a nonlinear modeling of the relation between RGB and luminance. On the other hand, to reduce the number of iterations needed, we use the direct model mentioned above: to set the RGB values of the next iteration we look at the differences between color prediction provided by the direct model for the current RGB values and desired color coordinates but looking separately at chromaticity and luminance following the same reasoning as for the initial point. As we will see from the experimental results, the method is accurate, efficient and robust. With respect to state of the art, method performance is specially good for low quality displays where physical assumptions made by other models do not hold completely., This work was supported by Generalitat Valenciana, Spain under grant IMaLeVICS CIAICO-2022-051 and Spanish Agencia Estatal de Investigacion under grant PID2022-140189OB-C21 and PID2023-152301OB-I00. The research of the fourth author is partially supported by Grant PID2021-124577NB-100 funded by MCIN/AEI/10.13039/501100011033 and by the "European Union NextGenerationEU/PRTR".