Excessive resident duty hours (RDH) are a recognized issue with implications for physician well-being and patient safety. A major component of the RDH concern is on-call duty. While considerable work has been done to reduce resident call workload, there is a paucity of research in optimizing resident call scheduling. Call coverage is scheduled manually rather than demand-based, which generally leads to over-scheduling to prevent a service gap. Machine learning (ML) has been widely applied in other industries to prevent such issues of a supply-demand mismatch. However, the healthcare field has been slow to adopt these innovations. As such, the aim of this study was to use ML models to 1) predict demand on orthopaedic surgery residents at a level I trauma centre and 2) identify variables key to demand prediction. Daily surgical handover emails over an eight year (2012-2019) period at a level I trauma centre were collected. The following data was used to calculate demand: spine call coverage, date, and number of operating rooms (ORs), traumas, admissions and consults completed. Various ML models (linear, tree-based and neural networks) were trained to predict the workload, with their results compared to the current scheduling approach. Quality of models was determined by using the area under the receiver operator curve (AUC) and accuracy of the predictions. The top ten most important variables were extracted from the most successful model. During training, the model with the highest AUC and accuracy was the multivariate adaptive regression splines (MARS) model, with an AUC of 0.78±0.03 and accuracy of 71.7%±3.1%. During testing, the model with the highest AUC and accuracy was the neural network model, with an AUC of 0.81 and accuracy of 73.7%. All models were better than the current approach, which had an AUC of 0.50 and accuracy of 50.1%. Key variables used by the neural network model were (descending order): spine call duty, year, weekday/weekend, month, and day of the week. This was the first study attempting to use ML to predict the service demand on orthopaedic surgery residents at a major level I trauma centre. Multiple ML models were shown to be more appropriate and accurate at predicting the demand on surgical residents as compared to the current scheduling approach. Future work should look to incorporate predictive models with optimization strategies to match scheduling with demand in order to improve resident well being and patient care.
Neuromuscular scoliosis patients face rates of major complications of up to 49%. Along with pre-operative risk reduction strategies (including nutritional and bone health optimization), intra-operative strategies to decrease blood loss and decrease surgical time may help mitigate these risks. A major contributor to blood loss and surgical time is the insertion of instrumentation which is challenging in neuromuscular patient given their abnormal vertebral and pelvic anatomy. Standard pre-operative radiographs provide minimal information regarding pedicle diameter, length, blocks to pedicle entry (e.g. iliac crest overhang), or iliac crest orientation. To minimize blood loss and surgical time, we developed an “ultra-low dose” CT protocol without sedation for neuromuscular patients. Our prospective quality improvement study aimed to determine: if ultra-low dose CT without sedation was feasible given the movement disorders in this population; what the radiation exposure was compared to standard pre-operative imaging; whether the images allowed accurate assessment of the anatomy and intra-operative navigation given the ultra-low dose and potential movement during the scan. Fifteen non-ambulatory surgical patients with neuromuscular scoliosis received the standard spine XR and an ultra-low dose CT scan. Charts were reviewed for etiology of neuromuscular scoliosis and medical co-morbidities. The CT protocol was a high-speed, high-pitch, tube-current modulated acquisition at a fixed tube voltage. Adaptive statistical iterative reconstruction was applied to soft-tissue and bone kernels to mitigate noise. Radiation dose was quantified using reported dose indices (computed tomography dose index (CTDIvol) and dose-length product (DLP)) and effective dose (E), calculated through Monte-Carlo simulation. Statistical analysis was completed using a paired student's T-test (α = 0.05). CT image quality was assessed for its use in preoperative planning and intraoperative navigation using 7D Surgical System Spine Module (7D Surgical, Toronto, Canada). Eight males and seven females were included in the study. Their average age (14±2 years old), preoperative Cobb angle (95±21 degrees), and kyphosis (60±18 degrees) were recorded. One patient was unable to undergo the ultra-low dose CT protocol without sedation due to a co-diagnosis of severe autism. The average XR radiation dose was 0.5±0.3 mSv. Variability in radiographic dose was due to a wide range in patient size, positioning (supine, sitting), number of views, imaging technique and body habitus. Associated CT radiation metrics were CTDIvol = 0.46±0.14 mGy, DLP = 26.2±8.1 mGy.cm and E = 0.6±0.2 mSv. CT radiation variability was due to body habitus and arm orientation. The radiation dose differences between radiographic and CT imaging were not statistically significant. All CT scans had adequate quality for preoperative assessment of pedicle diameter and orientation, obstacles impeding pedicle entry, S2-Alar screw orientation, and intra-operative navigation. “Ultra-low dose” CT scans without sedation were feasible in paediatric patients with neuromuscular scoliosis. The effective dose was similar between the standard preoperative spinal XR and “ultra-low dose” CT scans. The “ultra-low dose” CT scan allowed accurate assessment of the anatomy, aided in pre-operative planning, and allowed intra-operative navigation despite the movement disorders in this patient population.
Neuromuscular scoliosis patients face rates of major complications of up to 49%. Along with pre-operative risk reduction strategies (including nutritional and bone health optimization), intra-operative strategies to decrease blood loss and decrease surgical time may help mitigate these risks. A major contributor to blood loss and surgical time is the insertion of instrumentation which is challenging in neuromuscular patient given their abnormal vertebral and pelvic anatomy. Standard pre-operative radiographs provide minimal information regarding pedicle diameter, length, blocks to pedicle entry (e.g. iliac crest overhang), or iliac crest orientation. To minimize blood loss and surgical time, we developed an “ultra-low dose” CT protocol without sedation for neuromuscular patients. Our prospective quality improvement study aimed to determine: if ultra-low dose CT without sedation was feasible given the movement disorders in this population; what the radiation exposure was compared to standard pre-operative imaging; whether the images allowed accurate assessment of the anatomy and intra-operative navigation given the ultra-low dose and potential movement during the scan. Fifteen non-ambulatory surgical patients with neuromuscular scoliosis received the standard spine XR and an ultra-low dose CT scan. Charts were reviewed for etiology of neuromuscular scoliosis and medical co-morbidities. The CT protocol was a high-speed, high-pitch, tube-current modulated acquisition at a fixed tube voltage. Adaptive statistical iterative reconstruction was applied to soft-tissue and bone kernels to mitigate noise. Radiation dose was quantified using reported dose indices (computed tomography dose index (CTDIvol) and dose-length product (DLP)) and effective dose (E), calculated through Monte-Carlo simulation. Statistical analysis was completed using a paired student's T-test (α= 0.05). CT image quality was assessed for its use in preoperative planning and intraoperative navigation using 7D Surgical System Spine Module (7D Surgical, Toronto, Canada). Eight males and seven females were included in the study. Their average age (14±2 years old), preoperative Cobb angle (95±21 degrees), and kyphosis (60±18 degrees) were recorded. One patient was unable to undergo the ultra-low dose CT protocol without sedation due to a co-diagnosis of severe autism. The average XR radiation dose was 0.5±0.3 mSv. Variability in radiographic dose was due to a wide range in patient size, positioning (supine, sitting), number of views, imaging technique and body habitus. Associated CT radiation metrics were CTDIvol = 0.46±0.14 mGy, DLP = 26.2±8.1 mGy.cm and E = 0.6±0.2 mSv. CT radiation variability was due to body habitus and arm orientation. The radiation dose differences between radiographic and CT imaging were not statistically significant. All CT scans had adequate quality for preoperative assessment of pedicle diameter and orientation, obstacles impeding pedicle entry, S2-Alar screw orientation, and intra-operative navigation. “Ultra-low dose” CT scans without sedation were feasible in paediatric patients with neuromuscular scoliosis. The effective dose was similar between the standard preoperative spinal XR and “ultra-low dose” CT scans. The “ultra-low dose” CT scan allowed accurate assessment of the anatomy, aided in pre-operative planning, and allowed intra-operative navigation despite the movement disorders in this patient population.
