EP4677458A1 - Système et procédé combiné de sélection, d'annotation et d'entraînement via un modèle d'apprentissage automatique partagé - Google Patents
Système et procédé combiné de sélection, d'annotation et d'entraînement via un modèle d'apprentissage automatique partagéInfo
- Publication number
- EP4677458A1 EP4677458A1 EP24709410.5A EP24709410A EP4677458A1 EP 4677458 A1 EP4677458 A1 EP 4677458A1 EP 24709410 A EP24709410 A EP 24709410A EP 4677458 A1 EP4677458 A1 EP 4677458A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- data
- annotated
- machine learning
- learning model
- annotation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
- G06F18/2178—Validation; Performance evaluation; Active pattern learning techniques based on feedback of a supervisor
Definitions
- the invention relates to a combined system for selecting training data for a machine learning model, annotating training data for the model, training the learning model and predicting from the model, a method and an associated computer program product.
- the invention is particularly intended for the annotation of data related to the segmentation of medical images, for example from magnetic resonance imaging (MRI) or computed tomography (CT) for the identification of organs on each of the section images obtained by FIRM or CT in order to recreate a 3D model of the organ from these segmentations, among other applications.
- MRI magnetic resonance imaging
- CT computed tomography
- Image segmentation is a technique for associating each pixel or voxel of a 2D or 3D image with a label representing the object, texture or state, etc. corresponding to said pixel or voxel.
- Image segmentation is used in particular in a medical context to determine the organs, tissues, fluids, etc. visible on a 2D or 3D image.
- Such an image is for example obtained by MRI and makes it possible to form a 3D model of the targeted organ and for example of its environment and the tumor to be operated on.
- This creation of a 3D model requires the segmentation of the images obtained by FIRM.
- Image segmentation can be based on the use of a machine learning model that allows automatic or semi-automatic segmentation of images, respectively without or with the help of a user to make corrections.
- the machine learning model acts as a predictor that, by receiving as input an image to be segmented, provides a segmented image as output.
- Using a machine learning model requires prior training of the model using a large amount of annotated data, i.e. images in which the different segmentation classes have been determined and annotated/labeled. These annotated data are grouped into a dataset dedicated to training, validating and testing the implemented model.
- annotated data is an essential requirement for the supervised training of machine learning systems, which are defined by their ability to learn from this annotated data to make predictions or decisions.
- the predictor At the heart of a machine learning system is the predictor, which is trained by a machine learning model representing what has been learned by the system thanks to the annotated data.
- solutions to alleviate the data annotation problem have included, for example, semi-supervised, weakly supervised, self-supervised and unsupervised learning, zero-shot and one-shot learning, transfer learning and multitask learning. They can reduce or completely eliminate the need for human annotation.
- their applicability is limited: they are sensitive to errors, which makes them less suitable for complex, safety-critical, and high-variability problems, typically such as problems related to clinical and medical applications.
- These solutions offer less control over the quality of the results, generally require more complex algorithms and finer tuning, and can result in less accurate and less interpretable models than supervised learning.
- Supervised learning generally offers a simpler way to achieve better performance when there is sufficient annotated data, and therefore remains relevant and used by many current approaches.
- data annotation by a human expert remains useful and the search for a solution to speed up the annotation process is relevant. The human expert thus plays the role of an oracle for the machine learning system.
- the basic paradigm of machine learning has three main phases: a data annotation phase, a training phase of the predictive machine learning model, and an evaluation phase.
- the system can then return to the data annotation phase if necessary.
- the data annotation phase can use a classic annotation system such as 3D Slicer, which requires a lot of effort and time for a human oracle.
- this phase can be improved by using active learning to select the most useful unannotated data. This solution reduces the amount of data to be annotated while still providing relevant annotated data for training the machine learning model.
- Solutions propose to replace the classical annotation system with a dedicated annotation predictor.
- the annotation predictor suggests an annotation that the expert can validate or correct. This raises the question of training this annotation predictor.
- Existing systems train the annotation predictor once a sufficient amount of data has been annotated by the classical annotation system. They are suboptimal because the annotation predictor and the classical annotation system do not improve as data is annotated. One goal is therefore to exploit the data as it is annotated to train the annotation predictor to improve the annotation mechanism itself.
- the inventors therefore sought a solution to simplify the annotation task for the oracle by improving existing systems.
- the invention aims to provide a system, a method and a computer program product for selecting training data of a machine learning model, annotating training data of the model, training the model and predicting from the model.
- the invention also aims to provide, in at least one embodiment, a system for using a shared learning model to perform the selection of training data, and assist an oracle in the annotation of training data, while allowing the deployment of the shared model as a predictor at any time.
- the invention also aims to provide, in at least one embodiment, a system, a method and a computer program product allowing image segmentation, object detection and recognition, and/or image classification, more generally any type of task related to image processing, medical or not, or to computer vision, including, but not limited to classification, categorization, segmentation, semantic segmentation, instance segmentation, object detection, localization. Disclosure of the invention
- the invention relates to a combined system for selecting training data for a machine learning model and for annotating training data for the model, comprising: the pre-trained model with initial annotated training data, a first database comprising a set of unannotated data, a second database configured for storing annotated data, characterized in that the model is configured to: select, from among the set of unannotated data in the first database, a set of the most useful unannotated data to be annotated via an active learning method, automatically annotate the most useful unannotated data to provide automatically annotated data, receive, via an interactive annotation interface of the system associated with the machine learning model, at least one correction and/or validation of annotation of the data automatically annotated by an oracle to provide validated data, said validated data being stored in the second database, be retrained by the validated data of the second database for selection of the most useful unannotated data and for interactive annotation.
- a combined system according to the invention therefore makes it possible to manage different functions around a shared learning model so that the learning model benefits from this interaction between the different functions.
- the system allows on the one hand to select the training data by determining which unannotated data are the most useful for training thanks to the active learning implemented by the machine learning model. This allows to reduce the number of data to annotate by an oracle and allow continuous addition of new data to be annotated by prioritizing the unannotated data that contributes the most to improving the system's performance in the next training cycle.
- the system also provides a solution to assist a human oracle in the interactive annotation of this most useful data using this same machine learning model in order to simplify data annotation.
- the annotation is thus controlled and guided by the machine learning model.
- the more efficient the model the fewer interactions required by the oracle for data annotation, in particular because the number of corrections required by the oracle is reduced.
- the annotated data is also of better quality than pseudo-labeled data from self-training.
- the annotated data itself is used for retraining the model, in particular to assist in sorting the most useful non-annotated data to annotate, and to assist the oracle in annotating the non-annotated data.
- the machine learning model is better known as a Machine Learning model in English. The time spent annotating the data is thus reduced for the oracle, while retaining control of the annotation.
- Retraining broadly refers to several solutions for training the model with the entire annotated data set including the annotated data already used for training and the validated data, which can consist of: complete retraining, fine tuning or continuous learning.
- This retraining is carried out automatically or by a competent human expert to act on the machine learning model.
- the fusion of these functions allows on the one hand to limit the resources required for annotation, and on the other hand to improve the overall performance of the annotation system, by combining the learning and annotation channels in a single combined system through the joint use of active learning by the machine learning model and interactive annotation using the same machine learning model as an annotation assistant.
- Using a shared machine learning model therefore also makes it easier to manage the model and reduce technical and logistical interventions and delays, in particular by combining the annotation phase and the machine learning model training phase.
- Pre-training also makes it easier to deploy the machine learning model for its automatic annotation prediction function for unannotated data when the machine learning model is sufficiently trained, especially when the dataset that has been annotated by the oracle is large enough.
- Pre-training also makes it possible to quickly approach a deployable system that plays its role as a predictor during interactive annotation assisting the oracle, while allowing re-training with the new annotated data.
- pre-training requires only a minimal number of initial annotated training data because the oracle corrects the first predictions and the model improves quickly.
- Initial annotated data is also kept as test data to track the progress of re-training.
- the system allows a strong and crossed interaction between annotation and prediction: the selection of the data to be annotated is carried out with the aim of achieving a better prediction of the machine learning model and the prediction by the learning model allows assistance in the generation of annotations of non-annotated data.
- the active learning method selects data based on specific criteria that can be internal to the data, external to the data (metadata), dependent on the data already annotated, and/or the complexity of annotation, etc.
- the interactive annotation method is for example derived from prior art methods, for example a method derived from the methods described in one of the following publications:
- UI-Net Interactive Artificial Neural Networks for Iterative Image Segmentation Based on a User Model, in: Eurographics Workshop on Visual Computing for Biology and Medicine, The Eurographics Association.
- Prior art systems use different means for data selection, automatic annotation and validation/correction by the oracle. While some prior art systems already use machine learning models at each step of training, none of these prior art systems combine data selection, automatic annotation and validation/correction by the oracle for training a single model. Prior art systems typically use a model that is already fully trained at at least one of the steps, in particular for the annotation tool, and this already fully trained model has not benefited from the data improvements that the other functions within the scope of the invention would have provided.
- data annotation is typically performed by pre-trained tools, whether based on a machine learning model or not.
- Prior art machine learning models use an active learning method regardless of the annotation mechanism. If active learning is performed, it is not used for the improvement of the machine learning model that is used for automatic annotation and interactive annotation.
- the selection of the set of The most useful unannotated data to be annotated is performed via the machine learning model using one or more of the following selection criteria: one or more internal properties of the data, one or more external properties of the data, the uncertainty of prediction by the machine learning model, the representativeness of the annotated data and the unannotated data, a criterion of difficulty of annotation of the data.
- the system is configured for the annotation of new non-annotated data for at least one of the following applications: segmentation of new non-segmented or partially segmented images, detection and recognition of objects in new images, classification of new images, categorization of new images.
- the system is deployed for its final task and allows the annotation of new data not used for training, i.e. not used for pre-training or re-training.
- the invention also relates to a method for selecting training data of a machine learning model and annotating training data of the model, said model being pre-trained with initial annotated training data, comprising: a step of selection by the machine learning model and among the set of unannotated data, a set of the most useful unannotated data to be annotated via an active learning method, a step of automatic annotation by the machine learning model of the most useful unannotated data to provide automatically annotated data, a step of receiving at least one validation and/or one correction of annotation of the automatically annotated data by an oracle to provide validated data, a step of training the machine learning model by the validated data for the selection of the most useful unannotated data and for the interactive annotation.
- the invention also relates to a combined system, a combined method and a combined computer program product characterized in combination by all or part of the features mentioned above or below.
- FIG. 1 is a schematic view of a combined selection system training data of a machine learning model and annotation of training data of the model according to one embodiment of the invention.
- machine learning model is represented with a neural network symbol for illustrative purposes but other types of machine learning models can be used.
- Figure 1 schematically represents a combined system 10 for selecting training data of a machine learning model 12 and for annotating training data of the machine learning model 12 according to an embodiment of the invention.
- the system 10 comprises the machine learning model 12 and two databases for managing unannotated data and annotated data, in particular a first database 14 comprising a set of unannotated data and a second database 16 comprising a set of annotated data.
- the machine learning model 12 is pre-trained with initial annotated training data. Initial test data are also provided for testing the performance of the system after pre-training and each future training.
- MRI segmentation of the female pelvis on a dataset created by the applicant comprising ninety-seven MRI series representing three thousand and sixty-six slices in total; “CT” (computed tomography) type segmentation of the liver and pancreas on the “Medical Segmentation” datasets.
- CT computed tomography
- the model can be pre-trained on 12% (fifteen out of ninety-seven data volumes), 11% (fifteen out of one hundred and thirty-one data volumes), 12% (thirty-three out of two hundred and eighty-two volumes) and 3.3% (forty out of one thousand and three hundred volumes) of all data respectively to be able to efficiently produce quality annotations for the rest of the data using the system.
- Machine learning model 12 is for example a neural network, for example an existing encoder-decoder neural network architecture, for example a ResNet34 encoder described in the publication “He, K., Zhang, X., Ren, S., Sun, J., 2016. Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770-778.” and a generic decoder.
- a neural network for example an existing encoder-decoder neural network architecture, for example a ResNet34 encoder described in the publication “He, K., Zhang, X., Ren, S., Sun, J., 2016. Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770-778.” and a generic decoder.
- the system 10 when deployed, is able to automatically predict an annotation of the unannotated data, allowing a correction or verification by an oracle 18, which is a human user sufficiently competent for this correction or verification, in other words an human expert.
- Correction or verification by the oracle is particularly important in the medical context where the annotation of non-annotated data presents a challenge for a future therapeutic or surgical intervention, for example for a segmentation of medical images intended for the generation of 3D models of organs for assistance with surgery by augmented reality.
- the machine learning model 12 requires re-training to improve its performance. This training requires data that must be annotated by the human oracle 18.
- the first database 14 may include a large amount of unannotated data.
- the unannotated data is sorted to obtain a set 20 of the most useful unannotated data, thereby reducing the time required for the human oracle 18 to generate annotated training data.
- the learning model 12 thus includes as inputs all of the most useful unannotated data and interactions from the human oracle 18 for annotating the unannotated data.
- the human oracle 18 has access to a user interaction loop 22 connected by the machine learning model 12.
- the interaction loop 22 receives data automatically annotated by the machine learning model 12.
- the user interaction loop 22 makes it possible to take into account a correction and/or validation by the human oracle 18 of this annotated data, via a user interface of the interaction loop 22.
- the machine learning model 12 takes into account any corrections and proposes a new automatic annotation accordingly.
- machine learning model 12 is a method that can be called interactive annotation.
- a machine learning model system comprising such an interaction loop for taking into account annotations is for example described in the publication of the inventors of the present application: Mikhailov, L, Chauveau, B., Bourdel, N., Bartoli, A. (2022).
- a Deep Learning-Based Interactive Medical Image Segmentation Framework In: Wu, S., Shabestari, B., Xing, L. (eds) Applications of Medical Artificial Intelligence. AMAI 2022. Lecture Notes in Computer Science, vol 13540. Springer, Cham.
- the machine learning model 12 provides validated data 24 as output, which are stored in the second database 16.
- the new annotated data added to the second database 16 thus allow retraining of the machine learning model 12.
- An active learning method that can be used is for example a method based on uncertainty-based data selection through entropy calculation. This method is based in particular on the principle that a prediction with a high entropy linked to an uncertain prediction is representative of informative data and useful for training the model.
- Figure 2 schematically represents a combined method 100 of selecting training data of a machine learning model and annotating training data of the model according to an embodiment of the invention.
- the method 100 represents an iteration of data selection and data annotation that can be implemented by a system as described with reference to FIG. 1.
- the method can be implemented iteratively for the annotation of the unannotated data, until it is stopped or when all of the unannotated data are annotated.
- Each iteration ends with a retraining of the machine learning model from a training set comprising the new annotated data.
- the stopping of the iterations can in particular be carried out when the tests carried out with the test data give an error rate lower than a certain threshold, or when an expert human, for example the oracle, considers that the performance of the annotation by the machine learning model is sufficient.
- the method 100 comprises a step 104 of selection by the machine learning model and from the set of unannotated data 102, of a set 20 of the most useful unannotated data via an active learning method.
- the data 200 considered less useful are ignored and are not annotated at this iteration.
- the method 100 then comprises a step 106 of interactive annotation, comprising in particular a step of automatic annotation by the machine learning model of the most useful non-annotated data 20 to provide automatically annotated data, and a step of receiving at least one validation and/or one correction of annotation of the data automatically annotated by an oracle to provide validated data 24, which are annotated data and the annotation of which has been validated by the human oracle.
- the method finally comprises a step 108 of training the machine learning model using the validated data 24 for the selection of the most useful unannotated data and for interactive annotation.
- This improves the machine learning model for the next iteration, and the model is retrained at each iteration.
- This increases the number of annotated data used for training at each iteration, ensuring that the most useful data are annotated to optimize the annotation time by the oracle and increase the performance of the machine learning model.
- data selection maximizes the impact of each annotation on the performance of the machine learning model, allowing for rapid improvement in the performance of the machine learning model.
- the criteria for selecting the most useful data can for example be: related to the internal properties of the data, for example the type of image or the quality of the image; related to the external properties of the data, of the metadata type, for example a selection according to the age or sex of the patient from whom the data comes, the source of the data, etc.; related to the uncertainty of prediction by an inference of the machine learning model on all the non-annotated data; related to the representativeness, for example the images of which a characteristic is at a position far from the average; or personalized criteria, for example a pre-selection of data considered as the most difficult to annotate by a human oracle.
- the invention can also be used in contexts other than image segmentation, in particular object detection and recognition, and/or image classification, more generally any type of task related to image processing, medical or not, or to computer vision, including, but not limited to classification, categorization, segmentation, semantic segmentation, instance segmentation, object detection, localization.
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| FR2302198A FR3146529A1 (fr) | 2023-03-09 | 2023-03-09 | Système et procédé combiné de sélection, d’annotation et d’entraînement via un modèle d’apprentissage automatique partagé |
| PCT/EP2024/056148 WO2024184506A1 (fr) | 2023-03-09 | 2024-03-08 | Système et procédé combiné de sélection, d'annotation et d'entraînement via un modèle d'apprentissage automatique partagé |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| EP4677458A1 true EP4677458A1 (fr) | 2026-01-14 |
Family
ID=87137039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP24709410.5A Pending EP4677458A1 (fr) | 2023-03-09 | 2024-03-08 | Système et procédé combiné de sélection, d'annotation et d'entraînement via un modèle d'apprentissage automatique partagé |
Country Status (3)
| Country | Link |
|---|---|
| EP (1) | EP4677458A1 (fr) |
| FR (1) | FR3146529A1 (fr) |
| WO (1) | WO2024184506A1 (fr) |
-
2023
- 2023-03-09 FR FR2302198A patent/FR3146529A1/fr active Pending
-
2024
- 2024-03-08 WO PCT/EP2024/056148 patent/WO2024184506A1/fr not_active Ceased
- 2024-03-08 EP EP24709410.5A patent/EP4677458A1/fr active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| WO2024184506A1 (fr) | 2024-09-12 |
| FR3146529A1 (fr) | 2024-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11954610B2 (en) | Active surveillance and learning for machine learning model authoring and deployment | |
| Gorriz et al. | Cost-effective active learning for melanoma segmentation | |
| Singh et al. | Shifting to machine supervision: annotation-efficient semi and self-supervised learning for automatic medical image segmentation and classification | |
| Bakasa et al. | Stacked ensemble deep learning for pancreas cancer classification using extreme gradient boosting | |
| Araújo et al. | Artificial intelligence in healthcare applications targeting cancer diagnosis—part I: data structure, preprocessing and data organization | |
| Lu et al. | Deep learning-based image classification for integrating pathology and radiology in AI-assisted medical imaging | |
| Hanzala et al. | A hybrid approach for cervical cancer detection: Combining D-CNN, transfer learning, and ensemble models | |
| Vu et al. | Heritage image classification by convolution neural networks | |
| Cho et al. | Heterogeneous domain adaptation method for tomato leaf disease classification base on CycleGAN | |
| Wang et al. | An intelligent identification system combining image and DNA sequence methods for fruit flies with economic importance (Diptera: Tephritidae) | |
| Al Huda et al. | DeepMelaNet: Advancing melanoma stage classification in skin cancer diagnosis | |
| EP4677458A1 (fr) | Système et procédé combiné de sélection, d'annotation et d'entraînement via un modèle d'apprentissage automatique partagé | |
| Patel | Enhanced blood cell classification performance and conditional image generation with transformer based models | |
| Alqassab et al. | Identification of multiple ocular diseases using a hybrid quantum convolutional neural network with fundus images | |
| Wasson et al. | Image Quality Assessment: edge based entropy features estimation using Soft Computing Techniques | |
| Kothadiya et al. | Leveraging explainable AI in deep learning for brain tumor detection | |
| Ramesh et al. | EfficientNetB7-Based Deep Learning for Skin Lesion Classification | |
| Malagila Raphael et al. | Deep generative models in echocardiography: a scoping review of trends and future perspectives | |
| AlNasi et al. | Multi-task Weakly Supervised Intrapartum Ultrasound Measurements | |
| He | Deep learning approach for cell nuclear pore detection and quantification over high resolution 3D data | |
| Suresh et al. | MedicoCNN-GNN: A Hybrid Model with Integrated Attention Mechanism for Medical Image Classification | |
| Baby et al. | Multi-modal Representations for Fine-grained Multi-label Critical View of Safety Recognition | |
| Falissard | Deep learning methods in epidemiology and their applications to electronic health databases | |
| Namaemba | Maize Crop Pests and Diseases Classification Using Hybrid Models. | |
| US20250246010A1 (en) | Cross-modality pixel alignment and cell-to-cell registration across various imaging modalities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
| PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
| 17P | Request for examination filed |
Effective date: 20250905 |
|
| AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC ME MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40130940 Country of ref document: HK |