Predicting Mohs surgery complexity by applying machine learning to patient demographics and tumor characteristics

Abstract

Mohs micrographic surgery (MMS) is considered the gold standard for difficult‐to‐treat malignant skin tumors, whose incidence is on the rise. Currently, there are no agreed upon classifiers to predict complex MMS procedures. Such classifiers could enable better patient scheduling, reduce staff burnout and improve patient education. Our goal was to create an accessible and interpretable classifier(s) that would predict complex MMS procedures. A retrospective study applying machine learning models to a dataset of 8644 MMS procedures to predict complex wound reconstruction and number of MMS procedure stages. Each procedure record contained preoperative data on patient demographics, estimated clinical tumor size prior to surgery (mean diameter), tumor characteristics and tumor location, and postoperative procedure outcomes included the wound reconstruction technique and the number of MMS stages performed in order to achieve tumor‐free margins. For the number of stages complexity classification model, the area under the receiver operating characteristic curve (AUROC) was 0.79 (good performance), with model accuracy of 77%, sensitivity of 71%, specificity of 77%, positive prediction value (PPV) of 14% and negative prediction value (NPV) of 98%. The results for the wound reconstruction complexity classification model were 0.84 for the AUROC (excellent performance), with model accuracy of 75%, sensitivity of 72%, specificity of 76%, PPV of 39% and NPV of 93%. The ML models we created predict the complexity of the components that comprise the MMS procedure. Using the accessible and interpretable tool we provide online, clinicians can improve the management and well‐being of their patients. Study limitation is that models are based on data generated from a single surgeon.

Development and validation of a model to predict complex Mohs micrographic surgery in clinical practice: REGESMOSH scale

“…It is also important to consider that there is not an established definition of complex MMS surgery. We used three or more stages as a cut‐off point because it is the most frequently used cut‐off point in the stages number to define a complex closure 10,11,15 . Regarding the closure type, we did not differentiate the degree of difficulty or skill between flaps and grafts 33 considering that ‘complex’ refers to both surgical skills and the need for two surgical fields in the operating room, increasing the need for local anaesthesia or timing.…”

“…Our study shows factors that could help clinicians to predict complex surgery and complex closure as both factors will affect the time and resources needed in an intervention 18 . Although there is no consensus on what a complex surgery is, we used three or more stages as a cut‐off point, considering its frequency 10,11,15 and that each stage generally takes at least 30–45 min. Another important factor in the required time is the type of closure 12 .…”

“…Nevertheless, the scales did not correlate well with the number of stages or the surgery duration in an individual case 14 and its use has not been widespread 9 . Recently, a new algorithm has been proposed to predict the number of stages and complex wound reconstruction using patients' gender and age, tumour location, size, topography and characteristics 15 . However, all these models were validated in a single centre, raising doubts about the study external validity.…”

“…9 Recently, a new algorithm has been proposed to predict the number of stages and complex wound reconstruction using patients' gender and age, tumour location, size, topography and characteristics. 15 However, all these models were validated in a single centre, raising doubts about the study external validity.…”

Ultra-marginal Feature Importance

“…Marginal contribution feature importance (MCI) is a recent framework proposed in [7] to measure the marginal importance of a feature. MCI has been shown to work well in simulation studies as well as on real data [7,4,13]. However, there are two shortcomings of MCI.…”