Mobile Workflow in Computed Tomography of the Chest

Digitalization and the introduction of radiological information systems (RIS) and picture archiving and communication systems (PACS) allowed for digital image interpretation and established a linear workflow operating a computed tomography (CT) [1]. Advancements in hardware (tube, detector, or gantry) and software components (reconstruction algorithms, dose saving algorithms) evolved to faster acquisitions with higher resolutions at reduced radiation doses [2]. However, the CT acquisition sequence almost stayed the same until today: registration – positioning – planning – examination – release – reconstruction – archiving. Patient contact of the medical staff is limited to positioning and release of the patient in the examination room during this sequence. The majority of settings and adjustments are controlled from an operating console in a separate control room, where communication is limited to microphones and speakers. This separation between patients and medical staff strongly interferes with efforts to enforce compliance, especially in uncooperative patients. Injured patients, critically ill patients, demented patients or pediatric patients are especially at risk and may require assistance and surveillance during the entire examination, also during radiation [3]. Additional staff in the examination room, wrapped up with lead aprons, is still the only solution to overcome these situations. However, additional human resources are often limited, especially during nightshifts, weekends and holidays, in rural areas and in underdeveloped countries.

Mobile devices are already established in the consumer market and are also increasingly used for home medicine applications [4]. Since they are continuously improving in computing power and battery capacity, many implementations have recently been evaluated in radiologic departments, e.g. for patient briefing [5], diagnostic procedures [6], clinical knowledge assistance [7], case database management [8] or augmented reality for interventional procedures [9]. Recently a complete user interface application for mobile tablet devices has been presented for CT systems. A redesign of the examination workflow, bringing the radiographer closer to the patient, seems to be possible by integrating these mobile devices in the daily clinical routine. The aim of this study was to evaluate if the time spent together with the patient, which we consider as surrogate for patient contact and interaction, can be increased for radiographers using a mobile workflow to operate a CT of the chest in comparison to the conventional stationary console workflow. Patient proximity and patient experience are hardly measurable parameters. Therefore, camera surveillance and chronographic measurements were chosen for assessment of the surrogate. A prototype of x-ray protection cabin was additionally designed for this study to avoid the time consuming and inconvenient process of dressing and undressing lead aprons.

A total of 527 and 677 min were recorded by each camera for the mobile and conventional workflow. Mean age of the patients was 60.2 years in the study group and 60.3 years in the reference. In the study group 64% of patients were male compared to 60% in the reference. Indications for CT of the chest in the study/reference group were malignant disease in 15/19 cases, infectious disease in 16/13 cases, chronic interstitial lung disease in 10/10 cases and others in 6/9 cases. 3 wheelchair users were included in the mobile workflow compared to 9 in the conventional workflow. Detailed patient characteristics and radiation dose are listed in Table 1. A second radiographer was needed for assistance in 14 examinations of the study group and 10 examinations of the reference. Median age of the 10 different radiographers (2 male/ 8 female), who performed the examination, was 29 years (range 20–62 years) and the median clinical experience in CT was 3.5 years (range 0.5–18 years).

Radiographers’ time spent in the same room with the patient, which we consider as a surrogate for patient contact and quality in patient care, increased if a CT of the chest was examined with the help of mobile devices. Contact to the patient could be easily increased without a mobile workflow if the radiographers simply take more time for the procedures in the examination room. The drawback of this approach would be an extended total duration of the procedure and therefore a reduced number of patients that could be done per day. The main advantage of the mobile workflow found in this study is that the time spent together with the patient in the same room can be almost doubled without extending the total duration of the examination. Usage of an additional radiation protection cabin adds another 9% of the total duration to patient-vicinity in voice and visual contact.

It is well known that the efficiency of a CT system is mainly limited to the clinical workflow [12]. Several studies evaluated different techniques like intelligent scheduling or multiple radiographer workflows to increase the patient throughput of a CT system in literature [13] [14]. However, to our knowledge, no approach has yet been presented to redesign the conventional workflow sequence. Lin et al. were able to show that physicians’ time spent with patients is a determinant of patient satisfaction [15]. Although we didn’t assess patient satisfaction, our experience from this study is that the increased proximity of the radiographers to the patients is also beneficial for the compliance during the examination, especially in critical cases like excited or confused patients. It is also well known that children’s compliance often depends on their parents to be in the examination room [16]. The in-room radiation protection cabin presented in this study could be used for these cases as well to minimize the time of separation and to stay in voice and visual contact without increased radiation dose burden. Moreover, parents could also shortly walk in and out the cabin between the localizer and the tomogram to further reduce the separation from their children.

The radiographers’ increased freedom of movement and in-room solution of a radiation protection area presented in this study also opens up for additional concepts to further improve the cost-effectiveness of CT, even if not yet proven by our results. Future software versions of the mobile application could increase in functionality, so that in its strongest form mobile devices could completely replace the stationary console. If also a RIS was integrated on a mobile device, completely new room concepts would be feasible. Architectural planning could then discard the entire control room in favor of a radiation protection area in the examination room. Interventional procedures could benefit from applications like the one for image guidance that has already been reported by Hirata et al. [17], or as interventional suites in general without requiring other additional equipment than the tablet screen and mobile remote control.

Some limitations must be respected when interpreting our result: First, the mobile workflow was completely new to our radiographers while the conventional workflow has already been well established. Hence, it remains unclear if the rather little reduction of the total duration is due to the increased patient interaction, to the software or room concept in general, or to adaptation problems with the new situation. We expect that a learning effect and further improvements of the mobile workflow, like for example implementing a post-processing option on the tablet, could lead to additional time savings. Second, radiographers that operate a CT cannot be blinded to the procedure. Third, the camera surveillance was unable to differentiate between different radiographers and patients. The second supporting radiographer and patients that had to cross over from the preparation or changing room to the patient table imply a measurement bias. We assume this bias to be small because the frequency of a second assisting radiographer was comparable (30% and 20%) and patients’ interference should be identical and very short in both groups. The erroneous detection during transfer to and from the patient table only takes a few seconds, which we consider negligible in relation to the total duration. Moreover, this bias should not be represented in the time measurements since a second stopwatch was started whenever the second radiographer participated in the examination. Fourth, active time measurements using a stopwatch introduce their own errors. However, we consider the clinical collective large enough to overcome these rather slight inaccuracies. Fifth, we included all consecutive chest CT examinations in order to obtain a high sample size, with the drawback that we are unable to provide information about a special disease or degree of mobility. Sixth, contrast-enhanced and non-enhanced studies cannot be separated in the video evaluation. The knowledge from the time samplings that the contrast application accounts for approximately 10% of the total duration should be respected for the interpretation of the subjective maps. Anyway, we assume that the asymmetric ratio in the study (native/contrast = 3.7) and the reference (1.2) group rather favors an underestimation of area counts in the examination room for the mobile workflow. Seventh, counts in the control room decreased in all areas for the mobile workflow, although stable counts in the ‘RIS Area’ and the ‘Observer Area’ were expected. We attribute this effect to an inaccuracy of the video surveillance software. Persons moving on the edge of one area might overlap into adjacent areas. Eighth, the video surveillance software is limited to a cumulative assessment. Hence, we are unable to provide other statistical evaluation than the presented descriptive values.

Mobile workflows in CT of the chest transfers the radiographers’ interaction with the scanner from the control room into the examination room. There, radiographers’ freedom of movement and time spent in the same room with the patient is higher compared to the conventional setting, without affection of the total duration.