An initial learning experience of tracheal intubation with video laryngoscope: Experiences from a novice PGY

Development of an Artificial Intelligence-Based Image Recognition System for Time-Sequence Analysis of Tracheal Intubation

BACKGROUND

Total intubation time (TIT) is an objective indicator of tracheal intubation (TI) difficulties. However, large variations in TIT because of diverse initial and end targets make it difficult to compare studies. A video laryngoscope (VLS) can capture images during the TI process. By using artificial intelligence (AI) to detect airway structures, the start and end points can be freely selected, thus eliminating the inconsistencies. Further deconstructing the process and establishing time-sequence analysis may aid in gaining further understanding of the TI process.

METHODS

We developed a time-sequencing system for analyzing TI performed using a #3 Macintosh VLS. This system was established and validated on 30 easy TIs performed by specialists and validated using TI videos performed by a postgraduate-year (PGY) physician. Thirty easy intubation videos were selected from a cohort approved by our institutional review board (B-ER-107-088), and 6 targets were labeled: the lip, epiglottis, laryngopharynx, glottic opening, tube tip, and a black line on the endotracheal tube. We used 887 captured images to develop an AI model trained using You Only Look Once, Version 3 (YOLOv3). Seven cut points were selected for phase division. Seven experts selected the cut points. The expert cut points were used to validate the AI-identified cut points and time-sequence data. After the removal of the tube tip and laryngopharynx images, the durations between 5 identical cut points and sequentially identified the durations of 4 intubation phases, as well as TIT.

RESULTS

The average and total losses approached 0 within 150 cycles of model training for target identification. The identification rate for all cut points was 92.4% (194 of 210), which increased to 99.4% (179 of 180) after the removal of the tube tip target. The 4 phase durations and TIT calculated by the AI model and those from the expert exhibited strong Pearson correlation (phase I, r = 0.914; phase II, r = 0.868; phase III, r = 0.964; and phase IV, r = 0.949; TIT, r = 0.99; all P < .001). Similar findings were obtained for the PGY's observations (r > 0.95; P < .01).

CONCLUSIONS

YOLOv3 is a powerful tool for analyzing images recorded by VLS. By using AI to detect the airway structures, the start and end points can be freely selected, resolving the heterogeneity resulting from the inconsistencies in the TIT cut points across studies. Time-sequence analysis involving the deconstruction of VLS-recorded TI images into several phases should be conducted in further TI research.

A comparison of tracheal intubation using intubrite laryngoscope and conventional MAC laryngoscope: An open, prospective, crossover manikin study

BACKGROUND

New devices are more available in the pre-hospital environment operational theaters and emergency departments. One is an intubrite laryngoscope (INT) with Dual LED lighting that combines ultraviolet and white LED. The study aimed to compare the efficacy of endotracheal intubation using INT and conventional laryngoscope performed by inexperienced paramedics (paramedics students) and paramedics with experience in advanced airways management in full and limited accessibility settings.

METHODS

It was an open, prospective, crossover manikin study. Sixty paramedics and paramedic students were recruited. Participants were divided into 2 equal groups depending on their experience (n = 30). Experienced participants were further randomly divided into 2 groups (n = 15). Inexperienced participants were also randomly divided into 2 groups (n = 15). The criterion of inexperience was 5 or fewer intubation by any laryngoscope. Inexperience participants were asked to perform tracheal intubation in standard pre-hospital settings (without limited access to manikin) (scenario A) and difficult pre-hospital settings (limited access to manikin - narrow space between benches) (scenario B). Experience participants were asked to intubate manikin in difficult pre-hospital settings.

RESULTS

In the normal pre-hospital environment, the success rate after the first attempt was 56,7% for conventional laryngoscope and 66,7% for intubrite. However, the overall effectiveness of tracheal intubation using both laryngoscopes in 3 attempts was 90% for both devices. The successful rate of first attempt intubation in a difficult environment by inexperienced was 73,3% for INT and 50% for conventional laryngoscope. Overall effectiveness was 83,3% and 86,7% respectively. The successful rate of first attempt intubation in the experienced group was 86,7% with INT compared to 60% with a conventional laryngoscope in difficult settings. Overall effectiveness was 96,7% for both devices.

CONCLUSION

Intubrite provided better working conditions and make up for deficiencies in successful tracheal intubation by inexperienced participants in a normal and difficult environment. Tracheal intubation with intubrite was more effective in the experienced group. Tracheal intubation effectiveness with intubrite was also higher in the experienced group.

Intubation difficulty scale contributors and time delay in clinical practice

The intubation difficulty scale (IDS) includes 7 contributors that provide a comprehensive assessment of difficult intubation. However, the effect of each contributor is unclear, and the scale has not been revalidated recently and has not been validated in orient. This study determined the duration of successful intubation (DSI) for each of these 7 contributors.The patients were intubated by attending anesthesiologists. The duration and other data were recorded by 2 research assistants. Anaesthesiologists reported the IDS and their perceptions. A linear mixed-effects model with a DSI was constructed using IDS factors.In total, 1095 patients were enrolled. The average DSI was 23.9 ± 21.8 seconds (37.1% IDS = 0). All 7 factors were independently associated with duration, with the exception of vocal cord adduction. The best model was as follows: DSI (in seconds) = 15.2 + 31.1 (number of additional attempts) + 26.2 (number of additional operators) + 11.4 (number of alternative techniques) + 7.9 (increased lift force) + 4.9 (external laryngeal pressure) + 3.5 (Cormack grade 1). The mixed models were similar except for the regression coefficient for the number of alternative techniques that decreased from 11.4 to 6.9 seconds.We confirmed that each IDS contributor affects the DSI and validated a prediction model with 6 IDS contributors. This prediction model may facilitate the development of strategic plans for critical airway management. Furthermore, it could improve simulations and monitor learning progress and help provide valuable feedback.