Overcoming data scarcity in radiomics/radiogenomics using synthetic radiomic features.

PURPOSE: To evaluate the potential of synthetic radiomic data generation in addressing data scarcity in radiomics/radiogenomics models. METHODS: This study was conducted on a retrospectively collected cohort of 386 colorectal cancer patients (n = 2570 lesions) for whom matched contrast-enhanced CT images and gene TP53 mutational status were available. The full cohort data was divided into a training cohort (n = 2055 lesions) and an independent and fixed test set (n = 515 lesions). Differently sized training sets were subsampled from the training cohort to measure the impact of sample size on model performance and assess the added value of synthetic radiomic augmentation at different sizes. Five different tabular synthetic data generation models were used to generate synthetic radiomic data based on "real-world" radiomics data extracted from this cohort. The quality and reproducibility of the generated synthetic radiomic data were assessed. Synthetic radiomics were then combined with "real-world" radiomic training data to evaluate their impact on the predictive model's performance. RESULTS: A prediction model was generated using only "real-world" radiomic data, revealing the impact of data scarcity in this particular data set through a lack of predictive performance at low training sample numbers (n = 200, 400, 1000 lesions with average AUC = 0.52, 0.53, and 0.56 respectively, compared to 0.64 when using 2055 training lesions). Synthetic tabular data generation models created reproducible synthetic radiomic data with properties highly similar to "real-world" data (for n = 1000 lesions, average Chi-square = 0.932, average basic statistical correlation = 0.844). The integration of synthetic radiomic data consistently enhanced the performance of predictive models trained with small sample size sets (AUC enhanced by 9.6%, 11.3%, and 16.7% for models trained on n_samples = 200, 400, and 1000 lesions, respectively). In contrast, synthetic data generated from randomised/noisy radiomic data failed to enhance predictive performance underlining the requirement of true signal data to do so. CONCLUSION: Synthetic radiomic data, when combined with real radiomics, could enhance the performance of predictive models. Tabular synthetic data generation might help to overcome limitations in medical AI stemming from data scarcity.

Optimizing Wound Care after Surgery of the Head and Neck: A Review of Dressing Materials.

Wound healing is a complex biological process subject to complications that might jeopardize the patient's postoperative care. Appropriately approaching surgical wounds after head and neck surgery positively influences the quality and speed of wound healing and increases patient comfort. A large variety of dressing materials currently exist that allow the care of different types of wounds. Nevertheless, there is limited literature on the most suitable types of dressings after head and neck surgery. The objective of the present article is to review the most commonly used wound dressings, their benefits, indications, and disadvantages, and to provide a systematic approach for wound care within the head and neck. The Woundcare Consultant Society distinguishes wounds into three groups: black, yellow, and red. Each type of wound represents distinctive underlying pathophysiological processes with unique needs. Utilizing this classification along with the TIME model allows a proper characterization of wounds and the identification of potential healing barriers. This evidence-based and systematic approach can facilitate and guide the head and neck surgeon in selecting a wound dressing upon acknowledging their properties, which are herein reviewed and exemplified with representative cases.

Squamous Cell Carcinoma Antigen in the Follow-up of Patients With Head and Neck Cancer.

OBJECTIVE: to determine if the tumor marker squamous cell carcinoma antigen (SCC-Ag) observed over time may contribute to the early detection of recurrence, metastasis, and second primary tumors in the follow-up of patients with head and neck squamous cell carcinoma (HNSCC). STUDY DESIGN: A retrospective analysis of patients with HNSCC and at least one SCC-Ag measurement was conducted. Hazard ratios (HRs) were used to determine the correlation between SCC-Ag and an event. SETTING: patients with HNSCC, treated in the Antoni van Leeuwenhoek Hospital in The Netherlands between 2010 and 2020 were used for the analysis. METHODS: Data from 789 patients were used on event-free survival (EFS) with time-dependent Cox models. In addition to current (most recent) SCC-Ag (also dichotomized into high and low as done for clinical practice), average SCC-Ag and change between SCC-Ag measurements (delta SCC-Ag) were considered, using restricted cubic splines to explore nonlinear relationships. RESULTS: Dichotomized SCC-Ag values (HR = 3.01, 95% confidence interval [CI]: 2.17-4.18) and the delta SCC-Ag (HR = 1.15, 95% CI: 1.07-1.22) predicted EFS better than models using the cumulative average or current value of SCC-Ag, also after adjusting for tumor site, stage, age, and gender. A strong association was observed when using delta SCC-Ag as a linear predictor in the subgroup of oropharynx patients (HR = 4.88, 95% CI: 2.71-8.79). CONCLUSION: Dichotomized and delta SCC-Ag values can be important markers for EFS, during the follow-up of patients treated for HNSCC. These results were more evident in patients with oropharyngeal cancer.

Long-Term Esthetic Outcome of Different Treatment Modalities for Nontuberculous Mycobacterial Cervicofacial Lymphadenitis.

BACKGROUND: Nontuberculous mycobacterial (NTM) cervicofacial lymphadenitis is a rare infection which almost exclusively occurs in children, most commonly children 0-5 years old. It can leave scars in highly visible areas. The present study aimed to evaluate the long-term esthetic outcome of different treatment modalities for NTM cervicofacial lymphadenitis. METHODS: This retrospective cohort study included 92 participants with a history of bacteriologically proven NTM cervicofacial lymphadenitis. All patients were diagnosed at least 10 years prior and were aged >12 years upon enrollment. Based on standardized photographs, the scars were assessed by subjects with the Patient Scar Assessment Scale, and by five independent observers with the revised and weighted Observer Scar Assessment Scale. RESULTS: The mean age at initial presentation was 3,9 years and the mean follow-up time was 15.24 years. Initial treatments included surgical treatment (n = 53), antibiotic treatment (n = 29) and watchful waiting (n = 10). Subsequent surgery was performed in two patients, due to a recurrence after initial surgical treatment, and in 10 patients initially treated with antibiotic treatment or watchful waiting. Esthetic outcomes were statistically significantly better with initial surgery, compared to initial non-surgical treatment, based on patient scores of scar thickness, and based on observer scores of scar thickness, surface appearance, general appearance and the revised and weighted sum score of all assessment items. CONCLUSIONS: The long-term esthetic outcome of surgical treatment was superior to non-surgical treatment. These findings could facilitate the process of shared decision making. LEVEL OF EVIDENCE: Level III.

The Impact of Structured and Standardized Documentation on Documentation Quality; a Multicenter, Retrospective Study.

The reuse of healthcare data for various purposes will become increasingly important in the future. To enable the reuse of clinical data, structured and standardized documentation is conditional. However, the primary purpose of clinical documentation is to support high-quality patient care. Therefore, this study investigated the effect of increased structured and standardized documentation on the quality of notes in the Electronic Health Record. A multicenter, retrospective design was used to assess the difference in note quality between 144 unstructured and 144 structured notes. Independent reviewers measured note quality by scoring the notes with the Qnote instrument. This instrument rates all note elements independently using and results in a grand mean score on a 0-100 scale. The mean quality score for unstructured notes was 64.35 (95% CI 61.30-67.35). Structured and standardized documentation improved the Qnote quality score to 77.2 (95% CI 74.18-80.21), a 12.8 point difference (p < 0.001). Furthermore, results showed that structured notes were significantly longer than unstructured notes. Nevertheless, structured notes were more clear and concise. Structured documentation led to a significant increase in note quality. Moreover, considering the benefits of structured data recording in terms of data reuse, implementing structured and standardized documentation into the EHR is recommended.

Neoadjuvant immunotherapy with nivolumab and ipilimumab induces major pathological responses in patients with head and neck squamous cell carcinoma.

Surgery for locoregionally advanced head and neck squamous cell carcinoma (HNSCC) results in 30‒50% five-year overall survival. In IMCISION (NCT03003637), a non-randomized phase Ib/IIa trial, 32 HNSCC patients are treated with 2 doses (in weeks 1 and 3) of immune checkpoint blockade (ICB) using nivolumab (NIVO MONO, n = 6, phase Ib arm A) or nivolumab plus a single dose of ipilimumab (COMBO, n = 26, 6 in phase Ib arm B, and 20 in phase IIa) prior to surgery. Primary endpoints are feasibility to resect no later than week 6 (phase Ib) and primary tumor pathological response (phase IIa). Surgery is not delayed or suspended for any patient in phase Ib, meeting the primary endpoint. Grade 3‒4 immune-related adverse events are seen in 2 of 6 (33%) NIVO MONO and 10 of 26 (38%) total COMBO patients. Pathological response, defined as the %-change in primary tumor viable tumor cell percentage from baseline biopsy to on-treatment resection, is evaluable in 17/20 phase IIa patients and 29/32 total trial patients (6/6 NIVO MONO, 23/26 COMBO). We observe a major pathological response (MPR, 90‒100% response) in 35% of patients after COMBO ICB, both in phase IIa (6/17) and in the whole trial (8/23), meeting the phase IIa primary endpoint threshold of 10%. NIVO MONO's MPR rate is 17% (1/6). None of the MPR patients develop recurrent HSNCC during 24.0 months median postsurgical follow-up. FDG-PET-based total lesion glycolysis identifies MPR patients prior to surgery. A baseline AID/APOBEC-associated mutational profile and an on-treatment decrease in hypoxia RNA signature are observed in MPR patients. Our data indicate that neoadjuvant COMBO ICB is feasible and encouragingly efficacious in HNSCC.

Odontogenic keratocysts located in the buccal mucosa: A description of two cases and review of the literature.

Odontogenic keratocysts make up 4%-12% of all odontogenic cysts. Most cysts are sporadic but sometimes they arise in the context of basal cell nevus syndrome (Gorlin syndrome). Most odontogenic keratocysts arise in the posterior region of the mandible, but they can occur anywhere in the jaw. In rare instances, they are located peripherally in the gingiva. Even more rare, they are found in the soft tissues of the mouth. There have been a few case reports and small case series of such peripheral odontogenic keratocysts. Some controversy exists as to whether these truly represent a peripheral counterpart of the intraosseous odontogenic keratocysts and if their origin is at all odontogenic. We hereby present two cases of peripheral odontogenic keratocysts, both being located in the soft tissue of the buccal mucosa, and review the literature on peripheral odontogenic keratocysts.

Autologous Bone Is Inferior to Alloplastic Cranioplasties: Safety of Autograft and Allograft Materials for Cranioplasties, a Systematic Review.

BACKGROUND: Currently, various materials are routinely used for cranioplasty after decompressive craniectomy, each with their own features, potential benefits, and harms. OBJECTIVES: To systematically review available literature about safety (infection, resorption, and removal) of different materials used for cranioplasty for any indication. METHODS: A comprehensive search in MEDLINE, EMBASE, and the Cochrane library was performed for relevant studies published up to January 2017. Study quality was assessed according to the Cochrane Collaboration risk of bias assessment tool, and a set of 27 predetermined parameters was extracted by 2 investigators independently for further analysis. RESULTS: The search yielded 2 randomized, 14 prospective, and 212 retrospective studies, totaling 10,346 cranioplasties in which 1952 (18.9%) complications were reported in patients between 0 and 90 years old. Overall, study quality was low and heterogeneity was large. Graft infections and resorption were most prevalent: overall infection rate was 5.6%. Autologous cranioplasties showed an infection rate of 6.9% versus 5.0% in combined alloplastic materials, including poly(methyl methacrylate) with 7.8%. Resorption occurred almost exclusively in autologous cranioplasties (11.3%). The greatest removal rate was reported for autologous cranioplasties (overall: 10.4%), which was significantly greater than that of combined alloplastic materials (overall: 5.1%; risk difference = 0.052 [95% confidence interval: 0.039-0.066]; NNT = 19 [95% confidence interval: 15-25]). CONCLUSIONS: Available evidence on the safety of cranioplasty materials is limited due to a large diversity in study conduct, patients included, and outcomes reported. Autografts appear to carry a greater failure risk than allografts. Future publications concerning cranioplasties will benefit by a standardized reporting of surgical procedures, outcomes, and graft materials used.

Impact of Bone Volume Upon Condylar Activity in Patients With Unilateral Condylar Hyperplasia.

PURPOSE: Unilateral condylar hyperplasia or hyperactivity (UCH) is a bone overgrowth disorder affecting the mandible. The purpose of this study was to determine the relations among age, condylar bone structure, condylar bone volume, and condylar bone activity on single-photon emission computed tomographic (SPECT) scans in patients with UCH. MATERIALS AND METHODS: This study included 20 patients with a clinical presentation of progressive mandibular asymmetry and a positive bone SPECT scan. A bone SPECT-derived standardized uptake value (bSUV) for the condylar region was determined. All patients underwent condylectomy to arrest further progression of the disease. The resected condyles were scanned with a micro-computed tomographic scanner (18-µm resolution). Bone architectural parameters were calculated with routine morphometric software. RESULTS: The mean bSUV of the condyle on the affected side was 15.32 (standard deviation [SD], 8.98) compared with 9.85 (SD, 4.40) on the nonaffected side (P = .0007). For trabecular bone structure, there was a nonsignificant correlation between the SUV of the affected condyle and the measured bone volume fraction (r = 0.13; P = .58) and trabecular thickness (r = 0.03; P = .90). CONCLUSION: No meaningful relation was found between condylar bone volume fraction and condylar activity on bone scan; therefore, the impact of bone volume fraction on the results of bone scans is limited. The measured condylar activity on SPECT scan seems to be primarily a reflection of the remodeling rate of bone.

3-Dimensional CBCT analysis of mandibular asymmetry in unilateral condylar hyperplasia.

PURPOSE: Three-dimensional quantification of asymmetry in UCH has not been reported yet, but would be useful for diagnosing and evaluating the degree of deformity in this disease. It enables profound decision-making and timing of surgery. Unilateral condylar hyperplasia (UCH) can subjectively be classified in hemimandibular elongation (HE), hemimandibular hyperplasia (HH) and a combination of these two (hybrid form). The main purpose of this study was to quantify mandibular asymmetry in UCH patients with a reliable and reproducible method. Secondly, it was evaluated whether the existing classification can be confirmed. MATERIALS AND METHODS: 37 UCH-patients with progressive mandibular asymmetry, supported by a positive bone scan and/or such clinical progression that condylectomy was performed, were included in this retrospective study. A group of healthy subjects, matched for age and gender, was used as the control group. Cone-beam computed tomography (CBCT) scans were imported in Maxilim® software. Each mandibular half was divided into three skeletal segments (condyle, ramus, and body). Linear and volumetric measurements were calculated for these skeletal units on the affected and unaffected side, for both patients and controls. RESULTS: Significant differences between affected and unaffected sides in the patient group were found in condylar, ramus, and body segments for linear (p < 0.01) as well as for volumetric quantitative measurements (p < 0.0040). A mean linear difference between affected and unaffected sides in the condylar region of the UCH patient group was found of 3.6 mm (sd 2.9) versus 0.2 mm (sd 1.5) in controls. For volumetric measurements there was a mean difference between the left and right condyle of 718 mm3 (sd 638) in the patient group versus 8 mm3 (sd 225) difference in the control group. The condyle was the most affected segment. Differences between sides were significantly larger in the patient group than in the control group (p < 0.001). It was not possible to objectify differences between HE and HH. CONCLUSION: CBCT is a useful and accurate modality for quantification and evaluation of mandibular asymmetry in UCH. It enables objective monitoring. The existing classification in HE and HH could not be confirmed.

Interobserver variation of single-photon emission computed tomography bone scans in patients evaluated for unilateral condylar hyperactivity.

OBJECTIVE: This study assessed the quantitative evaluation of the region of interest (ROI) by mean and maximum pixel counts of single-photon emission computed tomography (SPECT) scans in patients with clinical suspicion of unilateral condylar hyperactivity (UCH); the interobserver reliability and the correlation of condylar activity with patient age was determined. METHODS: Two independent observers analyzed 67 bone SPECT scans. Maximum and mean activity values within an ROI on the mandibular condyle were analyzed; a cutoff value of 55% was used to determine the qualitative outcome. RESULTS: Excellent interobserver agreement was seen for both maximum (kappa 1.0) and mean activity analysis (kappa 0.94). Maximum and mean condylar activities were strongly correlated (r > 0.98). Maximum and mean condylar activity of the normal condyle decreased significantly with increasing age. CONCLUSIONS: Maximum and mean condylar activity levels were highly correlated in patients with mandibular asymmetry. An excellent interobserver agreement was found with either maximum or mean condylar activity.

Female predominance and effect of gender on unilateral condylar hyperplasia: a review and meta-analysis.

PURPOSE: The objective of this review and meta-analysis was to determine the existence and extent of the predominance of women among patients with unilateral condylar hyperplasia (UCH). Furthermore, we examined the laterality of UCH in women and men from international study populations. MATERIALS AND METHODS: We performed a review using a standardized search strategy and meta-analysis. RESULTS: We included 10 studies in this review, with a total of 275 UCH patients. The meta-analysis showed a clear predominance of female patients in the UCH study populations; the pooled proportion of female patients was 0.64 (95% confidence interval [CI], 0.58-0.70; n = 275 patients). The pooled percentage of female patients with left-sided UCH was 42% (95% CI, 34%-51%; n = 138), whereas 45% (95% CI, 33%-57%; n = 74) of male patients had left-sided UCH (P = .69 for female patients vs male patients). CONCLUSIONS: Female UCH patients outnumber male UCH patients in international study populations; therefore female gender may be considered a risk factor for UCH. We found no evidence of an association between laterality of UCH and gender: male patients and female patients have an almost equal distribution of left- and right-sided UCH.