Cascaded multi-scale convolutional encoder-decoders for breast mass segmentation in high-resolution mammograms.

This paper addresses breast mass segmentation from high-resolution mammograms. To cope with strong class imbalance, huge diversity of size, shape, texture and contour as well as limited receptive field, mass segmentation is achieved through a multi-scale cascade of deep convolutional encoder-decoders without any pre-detection scheme. Multi-scale information is integrated using auto-context to make long-range spatial context arising from lower scale impact training at higher resolution. The pipeline is trained end-to-end to benefit from simultaneous segmentation refinement performed at each level. It incorporates transfer learning and fine tuning from DDSM-CBIS to INbreast datasets to further improve mass delineations. The comprehensive evaluation provided for high-resolution INbreast images highlights promising model generalizability against standard encoder-decoder strategies.

Data hiding in homomorphically encrypted medical images for verifying their reliability in both encrypted and spatial domains.

In this paper, we propose a new scheme of data hiding of encrypted images for the purpose of verifying the reliability of an image into both encrypted and spatial domains. This scheme couples the Quantization Index Modulation (QIM) and the Paillier cryptosystem. It relies on the insertion into the image, before its encryption, of a predefined watermark, a "pre-watermark". Message insertion (resp. extraction) is conducted into (resp. from) the encrypted image using a modified version of QIM. It is the impact of this insertion process onto the "pre-watermark" that gives access to the message in the spatial domain, i.e. after the image has been decrypted. With our scheme, encryption/decryption processes are completely independent from message embedding/extraction. One does not need to know the encryption/decryption key for hiding a message into the encrypted image. Experiments conducted on ultrasound medical images show that the image distortion is very low while offering a high capacity that can support different watermarking based security objectives.

Computed tomography image source identification by discriminating CT-scanner image reconstruction process.

In this paper, we focus on the identification of the Computed Tomography (CT) scanner that has produced a CT image. To do so, we propose to discriminate CT-Scanner systems based on their reconstruction process, the footprint or the signature of which can be established based on the way they modify the intrinsic sensor noise of X-ray detectors. After having analyzed how the sensor noise is modified in the reconstruction process, we define a set of image features so as to serve as CT acquisition system footprint. These features are used to train a SVM based classifier. Experiments conducted on images issued from 15 different CT-Scanner models of 4 distinct manufacturers show it is possible to identify the origin of one CT image with high accuracy.

Secure Teleassistance towards endless medical litigations: identification of liabilities through a protocol using Joint Watermarking-Encryption Evidences.

Teleassistance is defined by the help provided through a telemedicine network by a medical practitioner to one other medical practitioner faced to a difficult case. One of the main limiting factors of its development is the fear of the practitioners to be involved in a litigation. In such a situation, the main issue is to determine as quick and as certain as possible if the damage is in relation with the tort of negligence and the liabilities of each involved physician. After a brief summary of the legal context, we present a protocol combining joint watermarking-encryption and a third party to enforce exchange traceability and therefore to bring valuable electronic evidence in case of teleassistance litigations.

Ontology-guided distortion control for robust-lossless database watermarking: application to inpatient hospital stay records.

In this paper, we propose a new semantic distortion control method for database watermarking. It is based on the identification of the semantic links that exist in-between attribute's values in tuples by means of an ontology. Such a database distortion control provides the capability for any watermarking scheme to avoid incoherent records and consequently ensures: i) the normal interpretation of watermarked data, i.e. introducing a watermark semantically imperceptible; ii) prevent the identification by an attacker of watermarked tuples. The solution we present herein successfully combines this semantic distortion control method with a robust lossless watermarking scheme. Experimental results conducted on a medical database of more than one half million of inpatient hospital stay records also show a non-negligible gain of performance in terms of robustness and database distortion.

Combination of watermarking and joint watermarking-decryption for reliability control and traceability of medical images.

In this paper, we propose a novel crypto-watermarking system for the purpose of verifying the reliability of images and tracing them, i.e. identifying the person at the origin of an illegal distribution. This system couples a common watermarking method, based on Quantization Index Modulation (QIM), and a joint watermarking-decryption (JWD) approach. At the emitter side, it allows the insertion of a watermark as a proof of reliability of the image before sending it encrypted; at the reception, another watermark, a proof of traceability, is embedded during the decryption process. The scheme we propose makes interoperate such a combination of watermarking approaches taking into account risks of interferences between embedded watermarks, allowing the access to both reliability and traceability proofs. Experimental results confirm the efficiency of our system, and demonstrate it can be used to identify the physician at the origin of a disclosure even if the image has been modified.

Robust lossless watermarking based on circular interpretation of bijective transformations for the protection of medical databases.

In this paper, we adapt the image lossless watermarking modulation proposed by De Vleeschouwer et al., based on the circular interpretation of bijective modulations, to the protection of medical relational databases. Our scheme modulates the numerical attributes of the database. It is suited for either copyright protection, integrity control or traitor tracing, being robust to most common database attacks, such as the addition and removal of tuples and the modification of attributes' values. Conducted experiments on a medical database of inpatient hospital stay records illustrate the overall performance of our method and its suitability to the requirements of the medical domain.

Medical image integrity control and forensics based on watermarking--approximating local modifications and identifying global image alterations.

In this paper we present a medical image integrity verification system that not only allows detecting and approximating malevolent local image alterations (e.g. removal or addition of findings) but is also capable to identify the nature of global image processing applied to the image (e.g. lossy compression, filtering …). For that purpose, we propose an image signature derived from the geometric moments of pixel blocks. Such a signature is computed over regions of interest of the image and then watermarked in regions of non interest. Image integrity analysis is conducted by comparing embedded and recomputed signatures. If any, local modifications are approximated through the determination of the parameters of the nearest generalized 2D Gaussian. Image moments are taken as image features and serve as inputs to one classifier we learned to discriminate the type of global image processing. Experimental results with both local and global modifications illustrate the overall performances of our approach.

A joint watermarking/encryption algorithm for verifying medical image integrity and authenticity in both encrypted and spatial domains.

In this paper, we propose a new joint watermarking/encryption algorithm for the purpose of verifying the reliability of medical images in both encrypted and spatial domains. It combines a substitutive watermarking algorithm, the quantization index modulation (QIM), with a block cipher algorithm, the Advanced Encryption Standard (AES), in CBC mode of operation. The proposed solution gives access to the outcomes of the image integrity and of its origins even though the image is stored encrypted. Experimental results achieved on 8 bits encoded Ultrasound images illustrate the overall performances of the proposed scheme. By making use of the AES block cipher in CBC mode, the proposed solution is compliant with or transparent to the DICOM standard.

Lossless watermarking of categorical attributes for verifying medical data base integrity.

In this article, we propose a new lossless or reversible watermarking approach that allows the embedding of a message within categorical data of relational database. The reversibility property of our scheme is achieved by adapting the well known histogram shifting modulation. Based on this algorithm we derive a system for verifying the integrity of the database content, it means detecting addition, removal or modification of any t-uples or attributes. Such a content integrity check is independent of the manner the database is stored or structured. We illustrate the overall capability of our method and its constraints of deployment considering one medical database of inpatient hospital stay records. Especially, we reversibly watermark ICD-10 diagnostic codes.

Blind forensics in medical imaging based on Tchebichef image moments.

In this paper, we present a blind forensic approach for the detection of global image modifications like filtering, lossy compression, scaling and so on. It is based on a new set of image features we proposed, called Histogram statistics of Reorganized Block-based Tchebichef moments (HRBT) features, and which are used as input of a set of classifiers we learned to discriminate tampered images from original ones. In this article, we compare the performances of our features with others proposed schemes from the literature in application to different medical image modalities (MRI, X-Ray …). Experimental results show that our HRBT features perform well and in some cases better than other features.

Reversible watermarking based on invariant image classification and dynamical error histogram shifting.

In this article, we present a novel reversible watermarking scheme. Its originality stands in identifying parts of the image that can be watermarked additively with the most adapted lossless modulation between: Pixel Histogram Shifting (PHS) or Dynamical Error Histogram Shifting (DEHS). This classification process makes use of a reference image derived from the image itself, a prediction of it, which has the property to be invariant to the watermark addition. In that way, watermark embedded and reader remain synchronized through this image of reference. DEHS is also an original contribution of this work. It shifts predict-errors between the image and its reference image taking care of the local specificities of the image, thus dynamically. Conducted experiments, on different medical image test sets issued from different modalities and some natural images, show that our method can insert more data with lower distortion than the most recent and efficient methods of the literature.

An additive and lossless watermarking method based on invariant image approximation and Haar wavelet transform.

In this article, we propose a new additive lossless watermarking scheme which identifies parts of the image that can be reversibly watermarked and conducts message embedding in the conventional Haar wavelet transform coefficients. Our approach makes use of an approximation of the image signal that is invariant to the watermark addition for classifying the image in order to avoid over/underflows. The method has been tested on different sets of medical images and some usual natural test images as Lena. Experimental result analysis conducted with respect to several aspects including data hiding capacity and image quality preservation, shows that our method is one of the most competitive existing lossless watermarking schemes in terms of high capacity and low distortion.

Comparison of some reversible watermarking methods in application to medical images.

Several reversible watermarking schemes have been proposed for images of sensitive content, like medical imaging, for which any modification may affect their interpretation. In this work, we distinguish these methods according to the way watermark insertion is conducted: additive and substitutive. Some of these approaches have been tested on different sets of medical images issued from three distinct modalities: Magnetic Resonance Images, Positron Emission Tomography and Ultrasound Imaging. Comparison analysis has been conducted with respect to several aspects including data hiding capacity and image quality preservation. Experimental results show different limitations which depend on the watermark approach but also on image modality specificities.

[Linking anonymous databases for national and international multicenter epidemiological studies: a cryptographic algorithm]. / Chaînage de bases de données anonymisées pour les études épidémiologiques multicentriques nationales et internationales : proposition d'un algorithme cryptographique.

BACKGROUND: Compiling individual records which come from different sources remains very important for multicenter epidemiological studies, but at the same time European directives or other national legislation concerning nominal data processing have to be respected. These legal aspects can be satisfied by implementing mechanisms that allow anonymization of patient data (such as hashing techniques). Moreover, for security reasons, official recommendations suggest using different cryptographic keys in combination with a cryptographic hash function for each study. Unfortunately, such an anonymization procedure is in contradiction with the common requirement in public health and biomedical research as it becomes almost impossible to link records from separate data collections where the same entity is not referenced in the same way. Solving this paradox by using methodology based on the combination of hashing and enciphering techniques is the main aim of this article. METHODS: The method relies on one of the best known hashing functions (the secure hash algorithm) to ensure the anonymity of personal information while providing greater resistance to dictionary attacks, combined with encryption techniques. The originality of the method relies on the way the combination of hashing and enciphering techniques is performed: like in asymmetric encryption, two keys are used but the private key depends on the patient's identity. RESULTS: The combination of hashing and enciphering techniques provides a great improvement in the overall security of the proposed scheme. CONCLUSION: This methodology makes the stored data available for use in the field of public health for the benefit of patients, while respecting legal security requirements.

Combining hashing and enciphering algorithms for epidemiological analysis of gathered data.

OBJECTIVES: Compiling individual records coming from different sources is necessary for multi-center studies. Legal aspects can be satisfied by implementing anonymization procedures. When using these procedures with a different key for each study it becomes almost impossible to link records from separate data collections. METHODS: The originality of the method relies on the way the combination of hashing and enciphering techniques is performed: like in asymmetric encryption, two keys are used but the private key depends on the patient's identity. RESULTS: The combination of hashing and enciphering techniques provides a great improvement in the overall security of the proposed scheme. CONCLUSION: This methodology makes stored data available for use in the field of public health, while respecting legal security requirements.

Medical image integrity control seeking into the detail of the tampering.

In this paper, we propose a system which aims at verifying integrity of medical images. It not only detects and localizes alterations, but also seeks into the details of the image modification to understand what occurred. For that latter purpose, we developed an image signature which allows our system to approximate modifications by a simple model, a door function of similar dimensions. This signature is partly based on a linear combination of the DCT coefficients of pixel blocks. Protection data is attached to the image by watermarking. Whence, image integrity verification is conducted by comparing this embedded data to the recomputed one from the observed image. Experimental results with malicious image modification illustrate the overall performances of our system.

Mixed reversible and RONI watermarking for medical image reliability protection.

By attaching image authenticity and integrity proofs directly at the pixels level of an image, watermarking can help to raise up medical image protection. However, because of the induced distortions, specific schemes like lossless data hiding or watermarking Regions of Non Interest (RONI) have been proposed to guarantee the preservation of the image interpretation. In this article, we propose an image reliability protection mechanism which combines advantages of both lossless and RONI approaches in order to provide a better and continuous protection. The proposed system was applied to magnetic resonance images of the head. Experimental results illustrate the overall functionality of the system.

A review of image watermarking applications in healthcare.

In this article, we focus on the complementary role of watermarking with respect to medical information security (integrity, authenticity ...) and management. We review sample cases where watermarking has been deployed, we conclude that watermarking has found a niche role in healthcare systems, as an instrument for protection of medical information, for secure sharing and handling of medical images. The concern of medical experts on the preservation of documents diagnostic integrity remains paramount.