Finding discriminative and interpretable patterns in sequences of surgical activities.

OBJECTIVE: Surgery is one of the riskiest and most important medical acts that is performed today. Understanding the ways in which surgeries are similar or different from each other is of major interest to understand and analyze surgical behaviors. This article addresses the issue of identifying discriminative patterns of surgical practice from recordings of surgeries. These recordings are sequences of low-level surgical activities representing the actions performed by surgeons during surgeries. MATERIALS AND METHOD: To discover patterns that are specific to a group of surgeries, we use the vector space model (VSM) which is originally an algebraic model for representing text documents. We split long sequences of surgical activities into subsequences of consecutive activities. We then compute the relative frequencies of these subsequences using the tf*idf framework and we use the Cosine similarity to classify the sequences. This process makes it possible to discover which patterns discriminate one set of surgeries recordings from another set. RESULTS: Experiments were performed on 40 neurosurgeries of anterior cervical discectomy (ACD). The results demonstrate that our method accurately identifies patterns that can discriminate between (1) locations where the surgery took place, (2) levels of expertise of surgeons (i.e., expert vs. intermediate) and even (3) individual surgeons who performed the intervention. We also show how the tf*idf weight vector can be used to both visualize the most interesting patterns and to highlight the parts of a given surgery that are the most interesting. CONCLUSIONS: Identifying patterns that discriminate groups of surgeon is a very important step in improving the understanding of surgical processes. The proposed method finds discriminative and interpretable patterns in sequences of surgical activities. Our approach provides intuitive results, as it identifies automatically the set of patterns explaining the differences between the groups.

Utility of a combined current procedural terminology and International Classification of Diseases, Ninth Revision, Clinical Modification code algorithm in classifying cervical spine surgery for degenerative changes.

STUDY DESIGN: Retrospective study. OBJECTIVE: To evaluate the sensitivity and specificity of a combined Current Procedural Terminology (CPT) and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) algorithm in defining cervical spine surgery in comparison to patient operative reports in the medical record. SUMMARY OF BACKGROUND DATA: Epidemiological studies of spine surgery often use ICD-9-CM billing codes in administrative databases to study trends and outcome of surgery. However, ICD-9-CM codes do not clearly identify specific surgical factors that may be related to outcome, such as instrumentation or number of levels treated. Previous studies have not investigated the sensitivity and specificity of a combined CPT and ICD-9-CM code algorithm for defining cervical spine surgical procedures. METHODS: We performed a retrospective study comparing the sensitivity and specificity of a combined CPT and ICD-9-CM code algorithm to the operative note, the gold standard, in a single academic center. We also compared the accuracy of our combined algorithm with our published ICD-9-CM-only algorithm. RESULTS: The combined algorithm has high sensitivity and specificity for defining cervical spine surgery, specific surgical procedures such as discectomy and fusion, and surgical approach. Compared to the ICD-9-CM-only algorithm, the combined algorithm significantly improves identification of discectomy, laminectomy, and fusion procedures and allows identification of specific procedures such as laminaplasty and instrumentation with high sensitivity and specificity. Identification of reoperations has low sensitivity and specificity, but identification of number of levels instrumented, fused, and decompressed has high specificity. CONCLUSION: The use of our combined CPT and ICD-9-CM algorithm to identify cervical spine surgery was highly sensitive and specific. For categories such as surgical approach, accuracy of our combined algorithm was similar to that of our ICD-9-CM-only algorithm. However, the combined algorithm improves sensitivity, and allows identification of procedures not defined by ICD-9-CM procedure codes, and number of levels instrumented and decompressed. The combined algorithm better defines cervical spine surgery and specific factors that may impact outcome and cost.

Transcutaneous lumbar diskectomy for internal disk derangement: a new indication.

BACKGROUND: Percutaneous diskectomy has been used effectively to treat lumbar disk herniation. The purpose of this review was to determine whether transcutaneous diskectomy is an effective surgical option for treating lumbar internal disk derangement. METHODS: Comprehensive retrospective review via personal interview included 103 consecutive patients available for long-term follow-up. For determining results as excellent, good, fair, or poor, the criteria included subjective relief of back and/or leg pain, ability to perform preinjury job functions, residual physical restrictions, use of medications, and need for subsequent surgery. Factors affecting results were statistically analyzed. RESULTS: The overall success rate was 83%, with no statistically significant differences in results based on sex, workers' compensation status, or levels of diskectomy. CONCLUSIONS: The favorable results from this series illustrate the high success rate of transcutaneous lumbar diskectomy in a heterogeneous patient population. Coupling the rate of success with a low rate of complications, transcutaneous lumbar diskectomy appears to be an effective, minimally invasive treatment for internal disk derangement.