公开(公告)号：US12118455B2

公开(公告)日：2024-10-15

申请号：US15965691

申请日：2018-04-27

Applicant: ARIZONA BOARD OF REGENTS ON BEHALF OF ARIZONA STATE UNIVERSITY

Inventor： Jianming Liang ,  Zongwei Zhou ,  Jae Shin

IPC: G06N3/08 ,  G06F18/21 ,  G06F18/214 ,  G06F18/2413 ,  G06F18/28 ,  G06N3/045 ,  G06N3/047 ,  G06V10/44 ,  G06V10/764 ,  G06V10/772 ,  G06V10/774 ,  G06V10/776 ,  G06V10/82

CPC classification number: G06N3/08 ,  G06F18/2148 ,  G06F18/217 ,  G06F18/2413 ,  G06F18/28 ,  G06N3/045 ,  G06N3/047 ,  G06V10/454 ,  G06V10/764 ,  G06V10/772 ,  G06V10/7747 ,  G06V10/776 ,  G06V10/82

Abstract: Systems for selecting candidates for labelling and use in training a convolutional neural network (CNN) are provided, the systems comprising: a memory device; and at least one hardware processor configured to: receive a plurality of input candidates, wherein each candidate includes a plurality of identically labelled patches; and for each of the plurality of candidates: determine a plurality of probabilities, each of the plurality of probabilities being a probability that a unique patch of the plurality of identically labelled patches of the candidate corresponds to a label using a pre-trained CNN; identify a subset of candidates of the plurality of input candidates, wherein the subset does not include all of the plurality of candidates, based on the determined probabilities; query an external source to label the subset of candidates to produce labelled candidates; and train the pre-trained CNN using the labelled candidates.

公开(公告)号：US12109061B2

公开(公告)日：2024-10-08

申请号：US17195694

申请日：2021-03-09

Applicant: SIEMENS HEALTHINEERS AG

Inventor： Lucian Mihai Itu ,  Tiziano Passerini ,  Saikiran Rapaka ,  Puneet Sharma ,  Chris Schwemmer ,  Max Schoebinger ,  Thomas Redel ,  Dorin Comaniciu

IPC: G06T7/11 ,  A61B5/00 ,  A61B5/026 ,  A61B6/00 ,  A61B6/03 ,  A61B6/50 ,  A61B8/06 ,  A61B8/08 ,  G06F18/21 ,  G06F18/22 ,  G06F18/2413 ,  G06T7/00 ,  G06V10/42 ,  G06V10/776 ,  G16H20/00 ,  G16H30/40 ,  G16H50/20 ,  G16H50/50 ,  A61B5/02 ,  A61B6/46 ,  A61B8/00 ,  G16H30/20

CPC classification number: A61B6/5217 ,  A61B5/026 ,  A61B5/7267 ,  A61B6/032 ,  A61B6/504 ,  A61B6/507 ,  A61B8/06 ,  A61B8/065 ,  A61B8/5223 ,  G06F18/217 ,  G06F18/22 ,  G06F18/2413 ,  G06T7/0012 ,  G06T7/11 ,  G06V10/42 ,  G06V10/776 ,  G16H20/00 ,  G16H30/40 ,  G16H50/20 ,  G16H50/50 ,  A61B5/02007 ,  A61B5/02028 ,  A61B5/0263 ,  A61B5/743 ,  A61B6/469 ,  A61B8/469 ,  A61B2576/00 ,  G06T2200/04 ,  G06T2207/10072 ,  G06T2207/10076 ,  G06T2207/20081 ,  G06T2207/30101 ,  G06T2207/30104 ,  G16H30/20

Abstract: In hemodynamic determination in medical imaging, the classifier is trained from synthetic data rather than relying on training data from other patients. A computer model (in silico) may be perturbed in many different ways to generate many different examples. The flow is calculated for each resulting example. A bench model (in vitro) may similarly be altered in many different ways. The flow is measured for each resulting example. The machine-learnt classifier uses features from medical scan data for a particular patient to estimate the blood flow based on mapping of features to flow learned from the synthetic data. Perturbations or alterations may account for therapy so that the machine-trained classifier may estimate the results of therapeutically altering a patient-specific input feature. Uncertainty may be handled by training the classifier to predict a distribution of possibilities given uncertain input distribution. Combinations of one or more of uncertainty, use of synthetic training data, and therapy prediction may be provided.

公开(公告)号：US12106214B2

公开(公告)日：2024-10-01

申请号：US17968085

申请日：2022-10-18

Applicant: SAMSUNG ELECTRONICS CO., LTD. ,  UNIVERSITAET ZUERICH

Inventor： Stefan Braun ,  Daniel Neil ,  Enea Ceolini ,  Jithendar Anumula ,  Shih-Chii Liu

IPC: G06N3/08 ,  G06F18/2413 ,  G06F18/25 ,  G06N3/04 ,  G06N3/0442 ,  G06N3/0455 ,  G06N3/0464 ,  G06N3/084 ,  G06V10/44 ,  G06V10/46 ,  G06V10/764 ,  G06V10/80 ,  G06V10/82 ,  G06V20/10 ,  G10L15/16 ,  G10L15/20 ,  G10L15/24

CPC classification number: G06N3/08 ,  G06F18/2413 ,  G06F18/256 ,  G06N3/04 ,  G06N3/0442 ,  G06N3/0455 ,  G06N3/0464 ,  G06N3/084 ,  G06V10/454 ,  G06V10/462 ,  G06V10/764 ,  G06V10/806 ,  G06V10/811 ,  G06V20/10 ,  G10L15/16 ,  G06V10/82 ,  G10L15/20 ,  G10L15/24

Abstract: A sensor transformation attention network (STAN) model including sensors configured to collect input signals, attention modules configured to calculate attention scores of feature vectors corresponding to the input signals, a merge module configured to calculate attention values of the attention scores, and generate a merged transformation vector based on the attention values and the feature vectors, and a task-specific module configured to classify the merged transformation vector is provided.

公开(公告)号：US12102441B2

公开(公告)日：2024-10-01

申请号：US17612486

申请日：2020-10-07

Applicant: INTROMEDIC CO., LTD.

Inventor： You Jin Kim

IPC: A61B5/00 ,  A61B1/00 ,  G06F18/22 ,  G06F18/2413 ,  G06N3/045 ,  G06N3/08 ,  G16H30/20 ,  G16H30/40 ,  G16H50/20 ,  G16H50/50

CPC classification number: A61B5/4255 ,  A61B1/00009 ,  A61B5/7264 ,  A61B5/7275 ,  A61B5/742 ,  G06F18/22 ,  G06F18/2413 ,  G06N3/045 ,  G16H30/40 ,  G16H50/20 ,  G06T2207/30028

Abstract: The present invention relates to a system for diagnosing small bowel cleanliness. The system may comprise: a similarity analysis unit for analyzing to select a representative image of similar small bowel images from among a plurality of small bowel images; an image classification unit for, when a series of a plurality of small bowel images in which cleanliness is to be diagnosed are received in a state where the plurality of small bowel images have been learned, classifying small bowel cleanliness according to scores by predicting the small bowel cleanliness by applying the representative image to a learning result; and a cleanliness diagnosis unit for calculating final small bowel cleanliness for the series of the plurality of small bowel images on the basis of a score for small bowel cleanliness of the representative image and the number of small bowel images similar to the representative image.

公开(公告)号：US12093838B2

公开(公告)日：2024-09-17

申请号：US17027688

申请日：2020-09-21

Applicant: International Business Machines Corporation

Inventor： Jing Xu ,  Si Er Han ,  Xue Ying Zhang ,  Steven George Barbee ,  Ji Hui Yang

IPC: G06N5/01 ,  G06F17/18 ,  G06F18/22 ,  G06F18/2413 ,  G06N7/01

CPC classification number: G06N5/01 ,  G06F17/18 ,  G06F18/22 ,  G06F18/2413 ,  G06N7/01

Abstract: Embodiments of the present disclosure relate to a method, system, and computer program product for efficient execution of a decision tree. According to the method, respective target values of a plurality of attributes of a target entity are obtained. Representations of a plurality of leaf nodes of a decision tree are obtained. Each of the representations indicates respective statistic values of a plurality of attributes of historical entities and a statistic prediction result determined from historical prediction results output at a respective one of the plurality of leaf nodes for the historical entities. Distance measures between the target entity and the plurality of leaf nodes are determined based on the target values and the statistic values indicated by the representations of the plurality of leaf nodes. A target prediction result for the target entity is determined based on the distance measures and the statistic prediction results of the historical entities.

公开(公告)号：US12080099B2

公开(公告)日：2024-09-03

申请号：US17394873

申请日：2021-08-05

Applicant: Ahmad Saleh

Inventor： Ahmad Saleh

IPC: G06K9/62 ,  G06F18/20 ,  G06F18/2413 ,  G06V40/16 ,  G06V40/20

CPC classification number: G06V40/172 ,  G06F18/2413 ,  G06F18/285 ,  G06V40/161 ,  G06V40/20

Abstract: A face mask detection system for detecting whether a user is wearing a mask at any given time using simple image recognition, configured to obtain an image of the user with the face mask, validate the image of the user wearing the face mask by an administrator (different from the user) in order to confirm validity of the image, and as a subsequent step use the validated image of the user with the face mask as a model image for benchmarking purposes. The model image is preferably a head profile of the user wearing a face mask. A face mask investigation unit compares the model image to an investigation image captured by an imaging system during an investigation process to determine whether the user in the investigation image is wearing the face mask based on whether an exact match is found between the model image and the investigation image.

公开(公告)号：US12080080B2

公开(公告)日：2024-09-03

申请号：US17373050

申请日：2021-07-12

Applicant: Toyota Motor Engineering & Manufacturing North America, Inc.

Inventor： Rohit Gupta ,  Ziran Wang ,  Yanbing Wang ,  Kyungtae Han ,  Prashant Tiwari

IPC: G06V20/59 ,  G06F18/214 ,  G06F18/2413 ,  G06N3/045 ,  G06N3/08 ,  G06V20/69 ,  G08B7/06

CPC classification number: G06V20/59 ,  G06F18/214 ,  G06F18/2413 ,  G06N3/045 ,  G06N3/08 ,  G06V20/698 ,  G08B7/06

Abstract: Systems, methods, and computer program products that are configured to identify or otherwise detect the presence of bacteria, classify the identified or detected bacteria, and also predict the growth of the classified bacteria on various touchable surfaces within a vehicle passenger cabin or compartment. Such systems, methods, and computer program products are configured to identify/detect, classify, and predict the presence and/or growth of bacteria, and transmit one or more alerts, warnings, and/or reports to vehicle owners, service providers, and/or occupants based on the identification/detection, classification, and prediction.

公开(公告)号：US12079311B2

公开(公告)日：2024-09-03

申请号：US17145123

申请日：2021-01-08

Applicant: Salesforce, Inc.

Inventor： Carlos Andres Esteva ,  Douwe Stefan van der Wal

IPC: G06T7/11 ,  G06F18/214 ,  G06F18/2413 ,  G06F18/40 ,  G06N3/08 ,  G06T7/00 ,  G06V10/25 ,  G06V10/40

CPC classification number: G06F18/2413 ,  G06F18/214 ,  G06F18/40 ,  G06N3/08 ,  G06T7/0012 ,  G06T7/11 ,  G06V10/25 ,  G06V10/40 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30024 ,  G06V2201/03

Abstract: An AI-enhanced data labeling tool assists a human operator in annotating image data. The tool may use a segmentation model to identify portions to be labeled. Initially, the operator manually annotates portions and once the operator has labeled a sufficient number of portions, a classifier is trained to predict labels for other portions. The predictions generated by the classifier are presented to the operator for approval or modification. The tool may also include an active learning model that recommends portions of the image data for the operator to annotate next. The active learning model may suggest one or more batches of portions based on the extent to which, once labeled, those batches will increase the diversity of the total set of labeled portions.

公开(公告)号：US12070693B2

公开(公告)日：2024-08-27

申请号：US18297140

申请日：2023-04-07

Applicant: LEGO A/S

Inventor： Marko Velic ,  Karsten Østergaard Noe ,  Jesper Mosegaard ,  Brian Bunch Christensen ,  Jens Rimestad

IPC: A63F13/65 ,  A63F13/213 ,  A63H33/08 ,  G06F18/2413 ,  G06F18/28 ,  G06N3/04 ,  G06N3/08 ,  G06N5/04 ,  G06V10/75 ,  G06V10/772 ,  G06V10/774 ,  G06V20/64 ,  G06V20/66

CPC classification number: A63F13/65 ,  A63F13/213 ,  A63H33/08 ,  G06F18/2413 ,  G06F18/28 ,  G06N3/04 ,  G06N3/08 ,  G06N5/04 ,  G06V10/751 ,  G06V10/772 ,  G06V10/774 ,  G06V20/64 ,  G06V20/66

Abstract: A recognition system for recognizing real-world toy objects from one or more images having an image capturing device and one or more processors. The processor implements a detection module, a recognition module, and a user experience module, and is configured to capture an image of a real-world scene, and detect one or more regions of interest in the image. The recognition system is configured to generate at least one part-image from the captured image, each part image including at least one of the one or more detected regions of interest, and feed the generated part-image to the recognition module. The recognition system is configured to recognize a real-world toy object in the part-image, the real-world toy object comprising at least one toy construction element, and provide a digital representation of the recognized real-world toy object.

公开(公告)号：US12050663B2

公开(公告)日：2024-07-30

申请号：US17490472

申请日：2021-09-30

Applicant: Toshiba Global Commerce Solutions Holdings Corporation

Inventor： Judith L. Atallah ,  J. Wacho Slaughter ,  Evgeny Shevtsov ,  Srija Ganguly

IPC: G06F18/2413 ,  G06N3/09 ,  G06Q20/20 ,  G06V10/764 ,  G07G1/00

CPC classification number: G06F18/2413 ,  G06Q20/202 ,  G06N3/09 ,  G06V10/764 ,  G07G1/0009

Abstract: This disclosure describes an automated process for training an ML model used by a computer vision system in a point of sale (POS) system to recognize a new item. Instead of relying on a manual process performed by a data scientist, the automated process can use images of a new (i.e., unknown) item captured at one or more POS systems to then retrain the ML model to recognize the new item. That is, the images of the item are used to retrain the ML model and to test the accuracy of the updated ML model. If the updated ML model can confidently identify the new item, the updated ML model is then used by the computer vision system to identify items at the POS system.