RESUMEN
This study examined the influence of familiarity and novelty on the mere exposure effect while manipulating the presentation of background information. We selected presentation stimuli that integrated cars and backgrounds based on the results of pilot studies. During the exposure phase, we displayed the stimuli successively for 3 seconds, manipulating the background information (same or different backgrounds with each presentation) and exposure frequency (3, 6, and 9 times). In the judgment phase, 18 participants judged the cars in terms of preference, familiarity, and novelty on a 7-point scale. As the number of stimulus presentations increased, the preference for the cars increased during the different background condition and decreased during the same background condition. This increased preference may be due to the increase in familiarity caused by the higher exposure frequency and novelty resulting from the background changes per exposure session. The rise in preference judgments was not seen when cars and backgrounds were presented independently. Therefore, the addition of novel features to each exposure session facilitated the mere exposure effect.
Asunto(s)
Conducta de Elección , Femenino , Humanos , Masculino , Fotograbar , Reconocimiento en Psicología , Adulto JovenRESUMEN
Numerous COVID-19 diagnostic imaging Artificial Intelligence (AI) studies exist. However, none of their models were of potential clinical use, primarily owing to methodological defects and the lack of implementation considerations for inference. In this study, all development processes of the deep-learning models are performed based on strict criteria of the "KAIZEN checklist", which is proposed based on previous AI development guidelines to overcome the deficiencies mentioned above. We develop and evaluate two binary-classification deep-learning models to triage COVID-19: a slice model examining a Computed Tomography (CT) slice to find COVID-19 lesions; a series model examining a series of CT images to find an infected patient. We collected 2,400,200 CT slices from twelve emergency centers in Japan. Area Under Curve (AUC) and accuracy were calculated for classification performance. The inference time of the system that includes these two models were measured. For validation data, the slice and series models recognized COVID-19 with AUCs and accuracies of 0.989 and 0.982, 95.9% and 93.0% respectively. For test data, the models' AUCs and accuracies were 0.958 and 0.953, 90.0% and 91.4% respectively. The average inference time per case was 2.83 s. Our deep-learning system realizes accuracy and inference speed high enough for practical use. The systems have already been implemented in four hospitals and eight are under progression. We released an application software and implementation code for free in a highly usable state to allow its use in Japan and globally.
Asunto(s)
COVID-19 , Aprendizaje Profundo , Humanos , COVID-19/diagnóstico por imagen , Inteligencia Artificial , Tomografía Computarizada por Rayos X/métodos , Programas Informáticos , Prueba de COVID-19RESUMEN
Isolated sphenoid sinus opacifications (ISSOs) are clinically important because they can lead to serious complications. However, some patients with ISSOs are asymptomatic, and not all patients are properly referred to the otolaryngology department. Because past studies of ISSOs focused only on patients who received treatment, in this study we selected ISSO cases based on radiology reports, then determined whether these patients had symptoms and were appropriately referred for specialty care. We conducted a retrospective analysis of data collected from patients who underwent computed tomography or magnetic resonance imaging from January 2007 to March 2017 at Osaka General Medical Center. We searched for the terms "sphenoid" or "sphenoidal" using F-REPORT to identify patients who had a sphenoid disease. We checked all selected images and diagnosed ISSOs. Examination of 1115 cases revealed 223 cases of ISSOs, of whom 167 (74.9%) were asymptomatic. We categorized patients with ISSOs into four groups: inflammation, mucocele, fungal diseases, and unclassifiable; the final category was used when edges were irregular or complete opacity was encountered. In the unclassifiable group, the majority of cases required otolaryngology consultation, but 37 of 47 unclassifiable patients did not have an otolaryngology visit. ISSOs are often identified by chance on imaging tests performed by non-otolaryngologists. However, our study revealed that many patients with ISSOs who should be treated by otolaryngologists were not referred to the otolaryngology department. Accordingly, it is important to promote awareness of the disease among other types of clinicians.