Search

Objective The risk of embolization to distal territory or to new territory in mechanical thrombectomy remains a major issue despite advancements in technological device. This condition can be caused by a large and firm dropped thrombus without passing through a guiding catheter during stent retriever or aspiration catheter withdrawal. This report introduced a novel technique referred to as retrograde angiography to detect dropped thrombus. Methods The retrograde angiography to detect dropped thrombus technique is a kind of retrograde angiography that consists of a contrast medium injection via a distal microcatheter and aspiration through an inflated balloon-guiding catheter. This method was used to detect dropped thrombus at the balloon-guiding catheter tip when back flow was blocked from the balloon-guiding catheter after stent retriever or aspiration catheter withdrawal. We retrospectively reviewed four consecutive patients who underwent the retrograde angiography to detect dropped thrombus technique during mechanical thrombectomy for acute ischemic stroke due to large vessel occlusion in the anterior circulation between January 2018 and January 2021. Results Three of four patients had dropped thrombus, which was diagnosed with the technique and retrieved completely with subsequent procedures while maintaining the balloon-guiding catheter inflated. None of the patients experienced embolization to distal territory/embolization to new territory, and a successful reperfusion was achieved in all four cases. Conclusions The retrograde angiography to detect dropped thrombus is a technique to detect a dropped thrombus at the balloon-guiding catheter tip and allows us to retrieve it with subsequent mechanical thrombectomy procedures while maintaining the balloon-guiding catheter inflated and it may be useful for reducing the risk of embolization to distal territory/embolization to new territory.

Dorsal arachnoid web (DAW) is a rare entity, which has been reported only in the thoracic spine. The authors report the first case of DAW developing in the cervical spine. A 78-year-old man with several-year progressive gait disturbance and bilateral lower-extremity numbness was referred to our hospital on the suspicion of a non-enhancing cystic cervical spinal tumor. Magnetic resonance imaging (MRI) showed a focal indentation along the dorsal surface of the spinal cord at C7 associated with widened cerebrospinal fluid (CSF) space and increased T2-weighted signals in the cord at C5–C7. DAW was suspected, but because of the atypical location for DAW, computed tomography (CT) myelogram was performed and demonstrated an incomplete blockage of the CSF flow at C7 with neither visible abnormal membranes nor a cyst formation. Intraoperative ultrasonography and operative findings revealed that two fluttering membranes disturbed the CSF flow. The pathology of the excised membranes was arachnoid tissues. DAW should be suspected based on the characteristic MRI findings even if the location is in the lower cervical spine. This case seems to support the theory that DAW may be an incomplete or disrupted formation of an arachnoid cyst.

In order to manage interruptions adequately, the prediction of an office worker’s uninterruptible duration is desired. We assumed three factors that may affect this uninterruptible duration: the type of work, the person’s level of concentration, and the frequency with which the person is disturbed by others. For features related to the type of work, we adopted category of using application and determined the ratio of key-to-mouse usage. The rates of keystroke and mouse operation and the application-switching frequency were selected to reflect a person’s concentration at work. A time of day was selected as an index which reflects the disturbance frequency. We then analyzed the relationship between these indices and the uninterruptible duration using 1200-h data. The results showed that, except for the time of day, a significant relationship exists between the uninterruptible duration and these indices. The combination of these indices appears promising for predicting the uninterruptible duration.

Experiments performed in controlled environments have revealed that interruptions occurring at application switching (AS), which correspond to task breakpoints in the computer field, are more subjectively acceptable and require shorter resumption lags (RL), which indicate the cognitive cost. Therefore, in this study, we investigate RLs in uncontrolled office environment. The interruptions at more acceptable ASs were expected to show shorter RLs, because the cognitive costs of the interruption are less. However, contrary to the expected result, the RLs were longer. This implies that factors other than cognitive cost may affect interruptibility in a realistic environment.

Feature selection is an important technique in pattern recognition. By removing features that have little or no discriminative information, it is possible to improve the predictive performance of classifiers and to reduce the measuring cost of features. In general, feature selection algorithms choose a common feature subset useful for all classes. However, in general, the most contributory feature subsets vary depending on classes relatively to the other classes. In this study, we propose a classifier as a decision tree in which each leaf corresponds to one class and an internal node classifies a sample to one of two class subsets. We also discuss classifier selection in each node.

In pattern recognition, feature selection is an important technique for reducing the measurement cost of features or for improving the performance of classifiers, or both. Removal of features with no discriminative information is effective for improving the precision of estimated parameters of parametric classifiers. Many feature selection algorithms choose a feature subset that is useful for all classes in common. However, the best feature subset for separating one group of classes from another may depend on groups. In this study, we investigate the effectiveness of choosing feature subsets depending on groups of classes (class-dependent features), and propose a classifier system that is built as a decision tree in which nodes have class-dependent feature subsets.