In-workflow artificial intelligence (ai)-enabled interruption handling for diagnostic radiology

#The following relates generally to the radiology arts, radiology workstation arts, radiology interpretation arts, radiology request allocation arts, and related arts.

#Radiology departments provide services for many (often most or all) clinical disciplines at a hospital or other medical institution. As such, a radiologist typically handles a heavy workload, and routinely deals with a wide range of different types of clinicians with varying knowledge of radiology. A radiologist often also has supervisory duties, for example providing support expertise for imaging technicians who operate magnetic resonance imaging (MRI) scanners, computed tomography (CT) scanners, positron emission tomography (PET) scanners, and/or imaging devices of other imaging modalities.

#One of the main measures of productivity in radiology is the volume of reads, where a “read” corresponding to reviewing, and preparing a radiology report on, a radiology examination of a single patient. While performing readings of radiology examinations may be the primary duty of a radiologist during a work shift, the radiologist is also expected to handle calls from physicians, imaging technicians, and others. These calls can be a substantial burden on the radiologist's time, and can break the radiologist's concentration when performing a complex reading. In some recent studies, radiologists were found to be interrupted every 4-12 minutes during regular business hours. See, e.g., A. Schemmel et al., “Radiology workflow disruptors: a detailed analysis,” Journal of the American College of Radiology, vol. 13, no. 10, pp. 1210-1214, 2016); R. M. Ratwani et al., “A human factors approach to understanding the types and sources of interruptions in radiology reading rooms,” Journal of the American College of Radiology, vol. 13, no. 9, pp. 1102-1105, 2016; T. Drew et al., “Quantifying the costs of interruption during diagnostic radiology interpretation using mobile eye-tracking glasses,” Journal of Medical Imaging, vol. 5, no. 3, p. 031406, 2018.). After hours, interruptions occur even more frequently, with on-call radiologists receive an average of 72 phone calls during a typical 12 hour shift (see, e.g., J. Y. John-Paul, A. P. Kansagra, and J. Mongan, “The radiologist's workflow environment: evaluation of disruptors and potential implications,” Journal of the American College of Radiology, vol. 11, no. 6, pp. 589-593, 2014.).

#The disclosure may take form in various components and arrangements of components, and in various steps and arrangements of steps. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the disclosure.

#FIG. 1 diagrammatically shows a workflow schedule monitoring system according to one aspect.

#FIG. 2 shows exemplary flow chart operations of the system of FIG. 1.

#Radiologists typically perform readings of radiology examinations in a dedicated “reading room” containing radiology workstations each having suitable high-resolution graphical display monitors for presenting radiology images, and running reading software for enabling the radiologist to review images (with zoom, pan, et cetera) and to write up (or dictate) radiology reports summarizing findings. Radiologist productivity is closely monitored, with radiologists expected to meet time and/or throughput goals for each work shift. At the same time, the radiologist must maintain a high level of accuracy and comprehensiveness in capturing salient medical findings from imaging examinations.

#Interruptions in the form of telephone calls, emails, text messages, and the like are a significant problem for radiologists. These can be from imaging technologists asking for clarification regarding a current imaging examination being performed; doctors calling to ask for clarification or other information about radiology reports they have received; other radiologists asking for advice; and so forth. By some estimates, a radiologist may receive around 6 interruptions every hour. Although frequent, these interruptions are typically not life-critical, as the radiologist is not an emergency care provider.

#The following discloses an interrupt handling system that intercepts and characterizes interruptions, prioritizes them, and to the extent practicable automatically resolves interruptions by way of a knowledge base (KB) or an artificial intelligence (AI) chat bot so as to reduce the number of interruptions that must be handled by a radiologist. The interruption handling system includes an interpreter that classifies an interruption, a scheduler that contextualizes the interruption, and a dispatcher that dispatches the interruption to the appropriate resolution entity, such as the aforementioned KB and/or AI chat bot, a call or message queue for a radiology specialist group (e.g. MM radiology experts), a call or message queue for a specific radiologist, and/or so forth.