Abstract: Data is received from a plurality of devices each having a same target part subject to failure. The received data is used to determine, for each of at least a subset of the plurality of devices, a part failure date on which the target part failed in that device. A set of features usable to predict failure of the target part is engineered, the set of features including one or more features that are not based on logged warning or error events. At least a subset of the data is labeled and aggregated over one or more days. The labeled and aggregated data is used to train a machine learning model configured to be used to predict failure of the target part in a device based on recent data from that device, including by computing from the data features corresponding to the programmatically engineered a set of features.

Abstract: Data is received from a plurality of devices each having a same target part subject to failure. The received data is used to determine, for each of at least a subset of the plurality of devices, a part failure date on which the target part failed in that device. A set of features usable to predict failure of the target part is engineered, the set of features including one or more features that are not based on logged warning or error events. At least a subset of the data is labeled and aggregated over one or more days. The labeled and aggregated data is used to train a machine learning model configured to be used to predict failure of the target part in a device based on recent data from that device, including by computing from the data features corresponding to the programmatically engineered a set of features.

Abstract: Techniques are disclosed to predict medical device failure based on operational log data. Log data associated with a plurality of devices comprising a population of devices each having a same target part subject to failure. For each of at least a subset of the plurality of devices replacement dates on which the target part was replaced in that device are determined. A set of logged event data with prescribed severity is extracted from the log data for said plurality of devices. A subset of the logged event data is identified as being associated with impending failure of the target part. The subset of the logged event data is transformed into a normalized form. The normalized subset of the logged event data is used to generate a failure prediction model to predict failure of the target part in a device based on the current event logs from that device.