Avoidable Factors Can Compromise Image-Guided Interventions
Avoidable Factors Can Compromise Image-Guided Interventions
Modern image-guided interventional and diagnostic medicine, such as minimally invasive surgery (MIS), endovascular surgery, endoscopy, natural orifice transluminal endoscopic surgery (NOTES), robotic surgery, interventional cardiology, and interventional neuroradiology and radiology, offers important advantages for the patient. These strategies are less invasive and thus less traumatic than comparable open surgeries. Image-guided medicine has also led to more day procedures and shorter hospital stays.
However, this very positive picture is incomplete; the benefits to patients are accrued at a considerable human-factor cost to the interventionist. Image-guided medicine requires physicians to work at the edge of their perceptual, cognitive, and psychomotor abilities. Their performance is easily stressed by seemingly benign human factors that could compromise patient safety. Physicians need to be aware of the research findings on factors that can compromise their performance so that they can manage or avoid them.
The skills required to perform minimally invasive image-guided procedures make unique demands on the operator. The surgeon or interventionist must operate remotely with long surgical instruments or catheters and wires that "fulcrum" against the body wall. During manipulations, the operator-held end and the working end of the instrument move in opposite directions; this causes a fundamental proprioceptive-visual conflict for novices. Figure 1a shows the observed difference in performance between 2 study groups who were required to perform a simple endoscopic task under normal laparoscopic imaging conditions and under Y-axis image inversion conditions.
Figure 1. Idealized graphs summarizing data from empirical studies on human factors that compromise performance of image-guided procedures.
Y-axis image inversion eliminates the appearance of the counterintuitive movement of the hand-held and working end of the instruments, thus eliminating the fulcrum effect. For scientific purposes, this configuration demonstrates and quantifies the considerable impediment that the fulcrum effect poses to the acquisition of technical skills.
In addition, practitioners of NOTES and robotic procedures have no haptic and tactile sensations from the instrument/tissue interaction. They must also operate while looking at a pixelated 2-dimensional image on a monitor, thus losing important cues that help in the judgment of depth of field. Figure 1b summarizes the results of a study that compared (1) task performance under normal full-binocular natural viewing conditions (looking directly at the task rather than an electronic and digital reconstruction of the task); (2) performance of the same task under the same conditions, but with 1 eye covered (monocular viewing); and (3) normal laparoscopic viewing conditions. The cognitive demands of the task on the participants were minimal in terms of intraperformance decision-making. The findings indicate the magnitude of performance degradation that resulted when the task was performed laparoscopically while looking at it on a monitor.
Performance degradation in MIS is substantially more than loss of binocularity, as indicated by the large difference between normal laparoscopic vision performance and full binocular performance in contrast to the relatively small (but statistically significant) difference between binocular and monocular viewing conditions. Contrary to commercial claims associated with supposed 3-dimensional camera systems, depth perception is a complex synthesis of information that is not fully restored by having the operator look at 2 slightly different images simultaneously. Surgeons who operate with fluoroscopic visually displayed information must do so with images that are degraded even further than those used in the standard experience of MIS surgeons and endoscopists.
Combined, these problems force the operator to work with considerably degraded, tactile, haptic and visual information, as well as a much greater demand on brain work compared with processing similar information under natural viewing and sensing conditions, as in traditional open surgery. These human factor constraints mean that minimally invasive image-guided procedures are more difficult to learn and to perform than traditional open surgical procedures and that surgeons are working at the limits of their perceptual, cognitive, and psychomotor capabilities. This has direct and quantifiable performance drawbacks.
Background
Modern image-guided interventional and diagnostic medicine, such as minimally invasive surgery (MIS), endovascular surgery, endoscopy, natural orifice transluminal endoscopic surgery (NOTES), robotic surgery, interventional cardiology, and interventional neuroradiology and radiology, offers important advantages for the patient. These strategies are less invasive and thus less traumatic than comparable open surgeries. Image-guided medicine has also led to more day procedures and shorter hospital stays.
However, this very positive picture is incomplete; the benefits to patients are accrued at a considerable human-factor cost to the interventionist. Image-guided medicine requires physicians to work at the edge of their perceptual, cognitive, and psychomotor abilities. Their performance is easily stressed by seemingly benign human factors that could compromise patient safety. Physicians need to be aware of the research findings on factors that can compromise their performance so that they can manage or avoid them.
Fundamental Human Factor Problems
The skills required to perform minimally invasive image-guided procedures make unique demands on the operator. The surgeon or interventionist must operate remotely with long surgical instruments or catheters and wires that "fulcrum" against the body wall. During manipulations, the operator-held end and the working end of the instrument move in opposite directions; this causes a fundamental proprioceptive-visual conflict for novices. Figure 1a shows the observed difference in performance between 2 study groups who were required to perform a simple endoscopic task under normal laparoscopic imaging conditions and under Y-axis image inversion conditions.
Figure 1. Idealized graphs summarizing data from empirical studies on human factors that compromise performance of image-guided procedures.
Y-axis image inversion eliminates the appearance of the counterintuitive movement of the hand-held and working end of the instruments, thus eliminating the fulcrum effect. For scientific purposes, this configuration demonstrates and quantifies the considerable impediment that the fulcrum effect poses to the acquisition of technical skills.
In addition, practitioners of NOTES and robotic procedures have no haptic and tactile sensations from the instrument/tissue interaction. They must also operate while looking at a pixelated 2-dimensional image on a monitor, thus losing important cues that help in the judgment of depth of field. Figure 1b summarizes the results of a study that compared (1) task performance under normal full-binocular natural viewing conditions (looking directly at the task rather than an electronic and digital reconstruction of the task); (2) performance of the same task under the same conditions, but with 1 eye covered (monocular viewing); and (3) normal laparoscopic viewing conditions. The cognitive demands of the task on the participants were minimal in terms of intraperformance decision-making. The findings indicate the magnitude of performance degradation that resulted when the task was performed laparoscopically while looking at it on a monitor.
Performance degradation in MIS is substantially more than loss of binocularity, as indicated by the large difference between normal laparoscopic vision performance and full binocular performance in contrast to the relatively small (but statistically significant) difference between binocular and monocular viewing conditions. Contrary to commercial claims associated with supposed 3-dimensional camera systems, depth perception is a complex synthesis of information that is not fully restored by having the operator look at 2 slightly different images simultaneously. Surgeons who operate with fluoroscopic visually displayed information must do so with images that are degraded even further than those used in the standard experience of MIS surgeons and endoscopists.
Combined, these problems force the operator to work with considerably degraded, tactile, haptic and visual information, as well as a much greater demand on brain work compared with processing similar information under natural viewing and sensing conditions, as in traditional open surgery. These human factor constraints mean that minimally invasive image-guided procedures are more difficult to learn and to perform than traditional open surgical procedures and that surgeons are working at the limits of their perceptual, cognitive, and psychomotor capabilities. This has direct and quantifiable performance drawbacks.
Source...