PATTERN LEARNING AND RECOGNITION DEVICE AND ASSOCIATED SYSTEM AND METHOD

    公开(公告)号:WO2023006805A1

    公开(公告)日:2023-02-02

    申请号:PCT/EP2022/071045

    申请日:2022-07-27

    Abstract: The present invention relates to a pattern learning and recognition device (16) comprising: - a training unit (22) adapted to train an oscillatory neural network, the training unit (22) being a part of a processor (28), - an oscillatory neural network unit (18), the oscillatory neural network unit (18) implementing a trained oscillatory neural network being adapted to output a pattern when an image is inputted, the oscillatory neural network unit (18) being a part of a programmable architecture (26), and - a controlling unit (20) adapted to control the oscillatory neural network unit (18) and the training unit (22), the controlling unit (20) being another part of the programmable architecture (26), the processor (28) and the programmable architecture (26) forming a system-on-chip (24).

    ALLELE-SPECIFIC GENOME EDITING OF THE NR2E3 MUTATION G56R

    公开(公告)号:WO2022180153A1

    公开(公告)日:2022-09-01

    申请号:PCT/EP2022/054629

    申请日:2022-02-24

    Abstract: Retinitis pigmentosa (RP) is an inherited retinal dystrophy that causes progressive vision loss. The second most common mutation causing autosomal dominant (ad) RP is the G56R mutation in NR2E3, a transcription factor essential for photoreceptor development. The G56R variant is exclusively responsible for all cases of NR2E3-associated adRP. Currently, there is no treatment for NR2E3-related, or other, adRP, but genome editing holds promise. In this study, the inventors developed a CRISPR/Cas strategy to specifically knockout the mutant G56R allele of NR2E3 and performed a proof-of-concept study in iPSC of an adRP patient. They demonstrate allele-specific knockout of the mutant G56R allele in the absence of off-target events. Furthermore, they validated this knockout strategy in an exogenous overexpression system. They showed for the first time that G56R iPSC, as well as G56R-CRISPR iPSC, can differentiate into NR2E3-expressing retinal organoids. Overall, they demonstrate that G56R allele-specific knockout by CRISPR/Cas could be a clinically relevant approach to treat NR2E3-associated adRP.Thus, the invention refers to a site-directed genetic engineering system for specifically editing an allele containing c.166G>A mutation in NR2E3 in the genome of an individual and its use for treating autosomal dominant retinitis pigmentosa.

    PROCEDE DE DETERMINATION AUTOMATIQUE DE PARAMETRES D'UN RESEAU DE NEURONES ARTIFICIELS ET MICROCONTROLEUR POUR LA MISE EN ŒUVRE DU PROCEDE

    公开(公告)号:WO2021105332A1

    公开(公告)日:2021-06-03

    申请号:PCT/EP2020/083594

    申请日:2020-11-27

    Inventor: MOHAMMADI, Bijan

    Abstract: Un aspect de l'invention concerne un procédé (200) de détermination automatique de paramètres d'un réseau de neurones, comportant les étapes suivantes : − Identification d'une géométrie ad hoc à une base de données (201); − Création d'un modèle comprenant une première couche (203); − Réalisation d'une itération (204) comportant les étapes suivantes : o Ajout d'une nouvelle couche dans le modèle; o Calcul d'un ensemble de coefficients synaptiques; o Pour chaque coefficient synaptique non nul, création de chaque scénario possible dans le modèle pour la synapse auquel est affecté le coefficient synaptique; o Pour chaque scénario, calcul d'une erreur d'apprentissage et d'une erreur de validation et arrêt du procédé si l'erreur d'apprentissage est supérieure à un seuil et/ou si l'erreur de validation croît; o Si le procédé n'est pas arrêté, réalisation d'une nouvelle itération (204) pour chaque scénario; les paramètres du réseau de neurones artificiels (100) correspondant aux paramètres du scénario du modèle R n'ayant pas vérifié la première condition pour lequel le procédé (200) a été arrêté.

    METHOD FOR SCREENING A SUBJECT FOR A CANCER
    9.
    发明申请

    公开(公告)号:WO2021023650A1

    公开(公告)日:2021-02-11

    申请号:PCT/EP2020/071637

    申请日:2020-07-31

    Inventor: THIERRY, Alain

    Abstract: The present invention relates to the field of cancer diagnostic. Here, the inventors observed by using conventional DSP sequencing method that size profile of double stranded DNA fragments obtained from cell free nucleic acids (cfDNA) may discriminate cfDNA from healthy and cancer derived subjects as previously observed (Jiang et al). Contrary to the prior art, the inventors determined specific double stranded DNA fragments or range of fragments and showed that these the number of these specific fragments are different between healthy and cancerous subjects. The number of double stranded DNA fragments as quantified from CfDNA are rather lower or higher when derived from healthy subject than from cancer subject. The invention provides description of calculation or biomarker from the identification of specific DNA fragments, specific ratios for different size or range of double stranded DNA fragment towards discriminating cancer to healthy plasma. The present invention relates to a method for screening a subject for a cancer by determining the level of double stranded DNA fragments in a sample.

    IN VITRO METHOD FOR ASSESSING THE RISK OF PROSTATE SIDE EFFECT AFTER TREATMENT BY IONIZING RADIATION

    公开(公告)号:WO2020178373A1

    公开(公告)日:2020-09-10

    申请号:PCT/EP2020/055813

    申请日:2020-03-05

    Abstract: The present invention relates to an in vitro method for assessing the risk of developing side effects after ionizing radiation treatment in a prostate cancer subject, comprising the steps of a)measuring radiation induced T-lymphocytes apoptosis in a sample of the subject; b)determining the presence of urinary toxicity in the patient prior to application of ionizing radiation, and optionally of at least one other clinical parameter, disease parameter or ionizing radiation treatment parameter from the patient, and c)combining the value of the at least one biochemical marker measured in step(a) and a value associated with the at least one clinical parameter, disease parameter or ionizing radiation treatment parameter determined in step (b) in a mathematical function to obtain an end-value.

Patent Agency Ranking