Generating a multimodal artificial intelligence model to differentiate benign and malignant follicular neoplasms of the thyroid: A proof-of-concept study.

BACKGROUND: Machine learning has been increasingly used to develop algorithms that can improve medical diagnostics and prognostication and has shown promise in improving the classification of thyroid ultrasound images. This proof-of-concept study aims to develop a multimodal machine-learning model to classify follicular carcinoma from adenoma. METHODS: This is a retrospective study of patients with follicular adenoma or carcinoma at a single institution between 2010 and 2022. Demographics, imaging, and perioperative variables were collected. The region of interest was annotated on ultrasound and used to perform radiomics analysis. Imaging features and clinical variables were then used to create a random forest classifier to predict malignancy. Leave-one-out cross-validation was conducted to evaluate classifier performance using the area under the receiver operating characteristic curve. RESULTS: Patients with follicular adenomas (n = 7) and carcinomas (n = 11) with complete imaging and perioperative data were included. A total of 910 features were extracted from each image. The t-distributed stochastic neighbor embedding method reduced the dimension to 2 primary represented components. The random forest classifier achieved an area under the receiver operating characteristic curve of 0.76 (clinical only), 0.29 (image only), and 0.79 (multimodal data). CONCLUSION: Our multimodal machine learning model demonstrates promising results in classifying follicular carcinoma from adenoma. This approach can potentially be applied in future studies to generate models for preoperative differentiation of follicular thyroid neoplasms.

Single site versus conventional laparoscopic appendectomy: some pain for no gain?

INTRODUCTION: The optimal laparoscopic appendectomy approach is not clear, comparing single site laparoscopic appendectomy (SILA) to conventional 3-port appendectomy (CLA). We investigated outcomes in pediatric patients comparing SILA to CLA: length of operation, length of stay, time to resumption of regular diet, follow up, rehospitalization, and cost. METHODS: Data was collected from children 1 to 18 years with appendectomy at Loma Linda University from 2018 to 2020, operated by two surgeons. Analysis utilized two-sample T, chi-squared, and Fisher's exact tests. RESULTS: Of 173 patients, 77 underwent SILA and 96 had CLA. There was no gender, age, or race difference between groups. Mean WBC was 17.5 × 103/mL in SILA group, compared to 15.3 × 103/mL in CLA group (P = 0.004). Operative time was 47.0 SILA compared to 49.5 minutes CLA (P = 0.269). Of SILA cases, 55.8% were simple appendicitis, while 53.3% of the CLA cases were simple (P = 0.857). Regular diet was resumed after 1.7 days in the SILA group, 1.1 days in CLA (P = 0.018). Length of stay was 2.9 days for SILA, 2.4 days for CLA (P = 0.144). Seven children required hospital readmission, 5 SILA and 2 CLA (P = 0.244). Five of the children who returned had intra-abdominal abscesses, of whom 4 had SILA. There was no difference in cost. CONCLUSIONS: The operative techniques had similar outcomes and operative times. There was a trend toward more intra-abdominal abscesses in the SILA group. Further study and longer follow up is needed to determine if there is an advantage to one laparoscopic approach over another.

The Impact of Primary Tumor Surgery on Survival in HER2 Positive Stage IV Breast Cancer Patients in the Current Era of Targeted Therapy.

OBJECTIVE: We sought to examine the impact of primary tumor resection on survival in HER2+ stage IV breast cancer patients in the era of HER2 targeted therapy. METHODS: We conducted a retrospective cohort study of women with HER2+ stage IV breast cancer in the National Cancer Database from 2010 to 2012 comparing those who did and did not undergo definitive breast surgery. RESULTS: Of 3231 patients, treatment included primary site surgery in 35.0%; chemo/targeted therapy in 89.4%; endocrine therapy in 37.7%; and radiation in 31.8%. Surgery was associated with Medicare/other government (OR 1.36, 95% CI 1.03-1.81) or private insurance (OR 1.93, 95% CI 1.53-2.42) versus none/Medicaid, radiation (OR 2.10, 95% CI 1.76-2.51), chemo/targeted therapy (OR 1.99, 95% CI 1.47-2.70), and endocrine therapy (OR 1.73, 95% CI 1.40-2.14). Non-Hispanic Black versus White patients (OR 0.68, 95% CI 0.53-0.87) were less likely to have surgery. Overall mortality was associated with insurance (Medicare/other government versus none/Medicaid, HR 0.36, p < 0.0001), receipt of chemo/targeted therapy (HR 0.76, p = 0.008), endocrine therapy (HR 0.70, p = 0.0006), and radiation therapy (HR 1.33, p = 0.0009), NH Black versus White race/ethnicity (HR 1.39, p = 0.002), visceral versus bone-only metastases (HR 1.44, p = 0.0003), and lowest versus highest income quartile (HR 1.36, p = 0.01). Propensity score analysis showed surgery was associated with improved survival versus no surgery (HR 0.56, 95% CI 0.40-0.77). CONCLUSIONS: Surgery of the primary site for metastatic HER2+ breast cancer is associated with improved overall survival in selected patients.

Impact of Breast Reconstruction on Time to Definitive Surgical Treatment, Adjuvant Therapy, and Breast Cancer Outcomes.

BACKGROUND: Delays in surgery and adjuvant treatment for breast cancer are associated with decreased survival. However, the time between diagnosis and surgery is rising, partly attributed to the added complexity of immediate breast reconstruction (IBR). We sought to investigate time to treatment and survival outcomes in breast cancer patients undergoing IBR. METHODS: We performed a retrospective review of 2004-2014 California Cancer Registry data for stage 0-III breast cancer patients undergoing mastectomy. Time to surgery, adjuvant systemic therapy and radiation therapy, propensity score, and covariate-adjusted overall mortality hazard ratios (HRs) were assessed by IBR status. RESULTS: Of 56,782 patients, 13,738 (24.2%) underwent IBR, with a median follow-up of 68.8 months. Median time between diagnosis and surgery was increased for patients undergoing IBR compared with those without {49 days (interquartile range [IQR] 34-73) vs. 35 days (IQR 21-56), p < 0.001}. IBR did not affect the interval from surgery to adjuvant chemotherapy or radiation, but prolonged time to endocrine therapy by 5 days (p = 0.014). Significantly lower survival was observed when time to surgery exceeded 120 days (vs. 0-30 days; HR 1.14 [1.02-1.28], p = 0.023), and improved survival with IBR (vs. without; HR 0.67 [0.61-0.74], p < 0.001). The benefit associated with reconstruction persisted for all age groups except age 80 + years, while surgical delay > 120 days demonstrated significantly lower survival in women < 60 years of age. CONCLUSIONS: While IBR delays time to definitive surgery, its use did not substantially affect time to adjuvant treatment or survival outcomes. Further research is ongoing to mitigate the effects of potential selection bias in favor of IBR.

Evaluation of Percutaneous First Annular Pulley Release: Efficacy and Complications in a Perfused Cadaveric Study.

PURPOSE: Trigger finger is the most common entrapment tendinopathy, with a lifetime risk of 2% to 3%. Open surgical release of the flexor tendon sheath is a commonly performed procedure associated with a high rate of success. Despite reported success rates of over 94%, percutaneous trigger finger release (PFTR) remains a controversial procedure because of the risk of iatrogenic digital neurovascular injury. This study aimed to evaluate the safety and efficacy of traditional percutaneous and ultrasound (US)-guided first annular (A1) pulley releases performed on a perfused cadaveric model. METHODS: First annular pulley releases were performed percutaneously using an 18-gauge needle in 155 digits (124 fingers and 31 thumbs) of un-embalmed cadavers with restored perfusion. A total of 45 digits were completed with US guidance and 110 digits were completed without it. Each digit was dissected and assessed regarding the amount of release as well as neurovascular, flexor tendon, and A2 pulley injury. RESULTS: Overall, 114 A1 pulleys were completely released (74%). There were 38 partial releases (24%) and 3 complete misses (2%). No significant flexor tendon injury was seen. Longitudinal scoring of the flexor tendon was found in 35 fingers (23%). There were no lacerations to digital nerves and one ulnar digital artery was partially lacerated (1%) in a middle finger with a partial flexion contracture that prevented appropriate hyperextension. The ultrasound-assisted and blind PTFR techniques had similar complete pulley release and injury rates. CONCLUSIONS: Both traditional and US-assisted percutaneous release of the A1 pulley can be performed for all fingers. Perfusion of cadaver digits enhances surgical simulation and evaluation of PTFR beyond those of previous cadaveric studies. The addition of vascular flow to the digits during percutaneous release allows for Doppler flow assessment of the neurovascular bundle and evaluation of vascular injury. CLINICAL RELEVANCE: Our cadaveric data align with those of published clinical investigations for percutaneous A1 pulley release.