A Scalable Radiomics- and Natural Language Processing-Based Machine Learning Pipeline to Distinguish Between Painful and Painless Thoracic Spinal Bone Metastases: Retrospective Algorithm Development and Validation Study.

BACKGROUND: The identification of objective pain biomarkers can contribute to an improved understanding of pain, as well as its prognosis and better management. Hence, it has the potential to improve the quality of life of patients with cancer. Artificial intelligence can aid in the extraction of objective pain biomarkers for patients with cancer with bone metastases (BMs). OBJECTIVE: This study aimed to develop and evaluate a scalable natural language processing (NLP)- and radiomics-based machine learning pipeline to differentiate between painless and painful BM lesions in simulation computed tomography (CT) images using imaging features (biomarkers) extracted from lesion center point-based regions of interest (ROIs). METHODS: Patients treated at our comprehensive cancer center who received palliative radiotherapy for thoracic spine BM between January 2016 and September 2019 were included in this retrospective study. Physician-reported pain scores were extracted automatically from radiation oncology consultation notes using an NLP pipeline. BM center points were manually pinpointed on CT images by radiation oncologists. Nested ROIs with various diameters were automatically delineated around these expert-identified BM center points, and radiomics features were extracted from each ROI. Synthetic Minority Oversampling Technique resampling, the Least Absolute Shrinkage And Selection Operator feature selection method, and various machine learning classifiers were evaluated using precision, recall, F1-score, and area under the receiver operating characteristic curve. RESULTS: Radiation therapy consultation notes and simulation CT images of 176 patients (mean age 66, SD 14 years; 95 males) with thoracic spine BM were included in this study. After BM center point identification, 107 radiomics features were extracted from each spherical ROI using pyradiomics. Data were divided into 70% and 30% training and hold-out test sets, respectively. In the test set, the accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve of our best performing model (neural network classifier on an ensemble ROI) were 0.82 (132/163), 0.59 (16/27), 0.85 (116/136), and 0.83, respectively. CONCLUSIONS: Our NLP- and radiomics-based machine learning pipeline was successful in differentiating between painful and painless BM lesions. It is intrinsically scalable by using NLP to extract pain scores from clinical notes and by requiring only center points to identify BM lesions in CT images.

Radiomics-based machine learning models to distinguish between metastatic and healthy bone using lesion-center-based geometric regions of interest.

Radiomics-based machine learning classifiers have shown potential for detecting bone metastases (BM) and for evaluating BM response to radiotherapy (RT). However, current radiomics models require large datasets of images with expert-segmented 3D regions of interest (ROIs). Full ROI segmentation is time consuming and oncologists often outline just RT treatment fields in clinical practice. This presents a challenge for real-world radiomics research. As such, a method that simplifies BM identification but does not compromise the power of radiomics is needed. The objective of this study was to investigate the feasibility of radiomics models for BM detection using lesion-center-based geometric ROIs. The planning-CT images of 170 patients with non-metastatic lung cancer and 189 patients with spinal BM were used. The point locations of 631 BM and 674 healthy bone (HB) regions were identified by experts. ROIs with various geometric shapes were centered and automatically delineated on the identified locations, and 107 radiomics features were extracted. Various feature selection methods and machine learning classifiers were evaluated. Our point-based radiomics pipeline was successful in differentiating BM from HB. Lesion-center-based segmentation approach greatly simplifies the process of preparing images for use in radiomics studies and avoids the bottleneck of full ROI segmentation.

Radiation-Induced Fibrosis in Patients with Head and Neck Cancer: A Review of Pathogenesis and Clinical Outcomes.

Radiotherapy-related fibrosis remains one of the most challenging treatment related side effects encountered by patients with head and neck cancer. Several established and ongoing novel therapies have been studied with paucity of data in how to best treat these patients. This review aims to provide researchers and health care providers with a comprehensive review on the presentation, etiology, and therapeutic options for this serious condition.

The role of pre-biopsy mpMRI in lymph node staging for prostate cancer.

INTRODUCTION: Prostate cancer (PCa) staging is an integral part in the management of prostate cancer. The gold standard for diagnosing lymph node invasion is a surgical lymphadenectomy, with no superior imaging modality available at the clinician's disposal. Our aim in this study is to identify if a pre-biopsy multiparametric MRI (mpMRI) can provide enough information about pelvic lymph nodes in intermediate and high risk PCa patients, and whether it can substitute further cross sectional imaging (CSI) modalities of the abdomen and pelvis in these risk categories. METHODS: Patients with intermediate and high risk prostate cancer were collected between January 2015 and June 2019, while excluding patients who did not undergo a pre-biopsy mpMRI or a CSI. Date regarding biopsy result, PSA, MRI results, CSI imaging results were collected. Using Statistical Package for the Social Sciences (SPSS) version 24.0, statistical analysis was conducted using the Cohen's Kappa agreement for comparison of mpMRI with CSI. McNemar's test and receiver operator curve (ROC) curve were used for comparison of sensitivity of both tests when comparing to the gold standard of lymphadenectomy. RESULTS: A total of 143 patients fit the inclusion criteria. We further stratified our patients into according to PSA level and Gleason score. Overall, agreement between mpMRI and all CSI was 0.857. When stratifying patients based on Gleason score and PSA, the higher the grade or PSA, the higher agreement between mpMRI and CSI. The sensitivity of mpMRI (73.7%) is similar to CSI (68.4%). When comparing CSI sensitivity to that of mpMRI, no significant difference was present by utilizing the McNemar test and very similar receiver operating characteristic curve. CONCLUSION: A pre-biopsy mpMRI can potentially substitute further cross sectional imaging in our cohort of patients. However, larger prospective studies are needed to confirm our findings.

Stereotactic Radiosurgery for Atrioventricular Node Ablation in Swine: A Study on Efficacy and Dosimetric Evaluation of Organs at Risk.

Introduction Stereotactic radiosurgery (SRS) delivered to arrhythmogenic foci within the heart is a promising treatment modality. We dosimetrically evaluated the radiation dose to the organs at risk of four swine that were successfully treated with linear-accelerator-based SRS for atrioventricular (AV) node ablation. Materials and methods Single-chamber pacemakers were implanted in four large white breed swine. Cardiac computed tomography simulation scans were performed to localize the AV node and organs at risk. SRS (35-40 Gy) was delivered to the AV node, and the pigs were followed up with pacemaker interrogations. One-sample t-tests were used to evaluate Dmax of great vessels, esophagus, and chest wall as compared to known normal tissue constraints as per RTOG 0631 and AAPM Task Group 101. Results All pigs had disturbances of AV conduction with progressive transition into complete heart block. Macroscopic and microscopic evaluation showed fibrosis in the AV node but did not reveal any changes in non-nodal cardiac tissue or vessels. The mean Dmax±SD (p-value) of the chest wall (14.7±3.3 (0.02)), esophagus (10.7±1.1 (<0.01)) superior vena cava (3.3±4.1 (<0.01)), right pulmonary artery (16.1±6.4 (<0.01)), right pulmonary vein (15.7± 5 (<0.01)), left pulmonary artery (11.1±1.7 (<0.01)) and left pulmonary vein (14.1±2.6 (<0.01)), and the inferior vena cava (33.68±1.6 (0.026)) were significantly below the normal tissue constraint cutoffs. Mean±SD (p-value) of the ascending aorta (19.4±16.1 (0.12)) was not significantly different than normal tissue constraint cutoffs. One swine model treated at 40 Gy had small area of hotspot in the ascending aorta (40.65 (0.4 cc)). Conclusion We have demonstrated in our swine models that SRS using 35-40 Gy can be done without exceeding known human normal tissue constraints to the chest wall, esophagus, and great vessels.

Development of a generalizable natural language processing pipeline to extract physician-reported pain from clinical reports: Generated using publicly-available datasets and tested on institutional clinical reports for cancer patients with bone metastases.

OBJECTIVE: The majority of cancer patients suffer from severe pain at the advanced stage of their illness. In most cases, cancer pain is underestimated by clinical staff and is not properly managed until it reaches a critical stage. Therefore, detecting and addressing cancer pain early can potentially improve the quality of life of cancer patients. The objective of this research project was to develop a generalizable Natural Language Processing (NLP) pipeline to find and classify physician-reported pain in the radiation oncology consultation notes of cancer patients with bone metastases. MATERIALS AND METHODS: The texts of 1249 publicly-available hospital discharge notes in the i2b2 database were used as a training and validation set. The MetaMap and NegEx algorithms were implemented for medical terms extraction. Sets of NLP rules were developed to score pain terms in each note. By averaging pain scores, each note was assigned to one of the three verbally-declared pain (VDP) labels, including no pain, pain, and no mention of pain. Without further training, the generalizability of our pipeline in scoring individual pain terms was tested independently using 30 hospital discharge notes from the MIMIC-III database and 30 consultation notes of cancer patients with bone metastasis from our institution's radiation oncology electronic health record. Finally, 150 notes from our institution were used to assess the pipeline's performance at assigning VDP. RESULTS: Our NLP pipeline successfully detected and quantified pain in the i2b2 summary notes with 93% overall precision and 92% overall recall. Testing on the MIMIC-III database achieved precision and recall of 91% and 86% respectively. The pipeline successfully detected pain with 89% precision and 82% recall on our institutional radiation oncology corpus. Finally, our pipeline assigned a VDP to each note in our institutional corpus with 84% and 82% precision and recall, respectively. CONCLUSION: Our NLP pipeline enables the detection and classification of physician-reported pain in our radiation oncology corpus. This portable and ready-to-use pipeline can be used to automatically extract and classify physician-reported pain from clinical notes where the pain is not otherwise documented through structured data entry.

Post-lumpectomy radiation therapy boost in breast cancer patients: evidence revisited.

PURPOSE: Radiation therapy is an integral part in the management of breast cancer after breast conservative surgery. In selected patients at high risk for local recurrence (LR), a boost radiation dose is commonly applied to the tumour bed. METHODS: We performed a review of the English literature using PubMed, Medline and Google Scholar for published manuscripts addressing the effect of boost radiation in breast cancer patients, focusing mainly on LR and overall survival (OS). RESULTS: A total of seven studies were included in our review. Most studies (6/7, 85.7%) showed a significant improvement in local control independent of age (hazard ratios ranging between 0.34 and 0.73), with the largest absolute benefit in younger patients. None of the studies, however, was able to demonstrate an improvement in OS. CONCLUSIONS: With lack of sufficient studies addressing the role of boost radiation, individualised treatment decisions are recommended, taking into account the risk factors for LR, including tumour biology. Real-life data are sorely needed to better assess the role of tumour bed boost in the contemporary era.

Predictors of Cardiac Sparing in Deep Inspiration Breath-Hold for Patients With Left Sided Breast Cancer.

Purpose: The purpose of this study was to evaluate patient-related non-dosimetric predictors of cardiac sparing with the use of deep inspiration breath-hold (DIBH) in patients with left-sided breast cancer undergoing irradiation (RT). Materials and Methods: We retrospectively reviewed charts and treatment plans of one-hundred and three patients with left-sided breast cancer. All patients had both free-breathing (FB) and DIBH (with body surface tracking) plans available. (MHD) and V4 (heart volume receiving at least 4 Gy) were extracted from dose volume histograms. Fisher's exact and Chi-square tests were used to identify predictors of reductions in MHD and V4 after DIBH. Results: One-hundred and three patients were identified and most underwent mastectomy. MHD and V4 decreased significantly in DIBH plans (0.74 ± 0.25 Gy vs. 1.72 ± 0.98 Gy, p < 0.0001 for MHD; 4 ± 4.98 cc vs. 20.79 ± 18.2 cc, p < 0.0001 for V4). Body mass index (BMI), smoking and timing of CT simulation (spring/winter vs. summer/fall) were significant predictors of reduction in MHD whereas BMI, field size, chemotherapy, axillary dissection, and timing of CT simulation predicted reduction in V4. On multivariate analysis, BMI, and timing of CT simulation remained significant predictors of the heart-sparing effect of DIBH. Conclusions: In the setting of limited resources, identifying patients who will benefit the most from DIBH is extremely important. Prior studies have identified multiple dosimetric predictors of cardiac sparing and hereby we identified new non-dosimetric factors such as BMI and timing of treatments.

Evidence and temporality of the obesity paradox in coronary bypass surgery: an analysis of cause-specific mortality.

OBJECTIVES: We evaluated the presence of an 'obesity paradox' in coronary artery bypass grafting (CABG) patients, determined its time course and ascertained whether it is associated with improved cardiovascular (CV) survival versus non-CV survival. METHODS: A retrospective analysis of 3 prospectively collected databases was conducted. A fifteen-year Kaplan-Meier analysis in 7091 CABG patients was performed and repeated in 5 body mass index [BMI (kg/m2)] cohorts [Normal (18.5-24.99 kg/m2), Overweight (25-29.99 kg/m2), Obese I (30-34.99 kg/m2), Obese II (35-39.99 kg/m2) and Obese III (≥40 kg/m2)]. Mortality hazard ratios {HR [95% confidence interval (CI)]} were derived using comprehensive multivariable competing risk Cox regression, accounting for BMI categories for overall (0-15), Early (0-1), Intermediate (1-8) and Late (8-15) postoperative years, to relax the proportional hazards assumption. The regression was repeated using BMI as a continuous variable. Mortality was classified into any, CV and non-CV. RESULTS: Obese patients were younger with more comorbidities. Fifteen-year survival was improved in the Overweight and Obese I groups (P < 0.001). Adjusted 15-year mortality was reduced in the Overweight [HR (95% CI) = 0.88 (0.79-0.98)] and Obese I [HR = 0.88 (0.78-0.99)] groups driven by improved CV and non-CV survival. This trend was noted in the early (Overweight) and intermediate postoperative periods (Overweight and Obese I) with no significance in the late period. Higher mortality in the Obese III [HR = 1.28 (1.06-1.55)] group was driven by a decreased CV survival. Using BMI as a continuous variable, a BMI of 29 kg/m2 was associated with optimal survival. CONCLUSIONS: We identified a protective partial obesity paradox in the early and intermediate postoperative periods among Overweight and mildly obese (Obese I) patients with improved CV and non-CV survival. The morbidly obese (the Obese III group) had higher early and late CV mortality.

Randomized, Prospective, Open-label Phase III Trial Comparing Mebo Ointment With Biafine Cream for the Management of Acute Dermatitis During Radiotherapy for Breast Cancer.

PURPOSE: Acute radiation dermatitis is a common side-effect of radiotherapy in breast cancer and has a profound impact on patients' quality of life, due to pain and discomfort. The aim of this study is to compare the effect of ß-sitosterol (Mebo) ointment to trolamine (Biafine) cream for the prevention and treatment of radiation dermatitis in breast cancer patients receiving adjuvant radiation therapy. MATERIALS AND METHODS: This is a prospective open-label randomized phase III study developed to assess the efficacy of 2 topical agents used for management of acute radiation dermatitis. Female breast cancer patients who needed a course of radiation therapy in our institution were enrolled and randomized into 2 groups 1 with Mebo ointment and 1 with Biafine cream. Both medications were applied twice per day during the whole period of treatment and skin reactions and related symptoms were assessed weekly during the entire course. Grading of skin reactions was done according to the Radiation Therapy Oncology Group grading system. RESULTS: Between September 2015 and May 2017, a total of 161 patients were recruited for this trial. Mean age was similar for both groups (50.19±12.57 vs. 51.73±11.23, respectively, P=0.41). All other patients and treatment characteristics were similar in both groups, except for the use of boost (82.7% in the Biafine group vs. 36.7% in Mebo group, P=0.012). Analysis was done for reactions recorded before the beginning of the boost and for the entire course including the boost. Using univariate and multivariate analysis, there was no significant difference in grades 2 and 3 dermatitis between the 2 groups. However, the incidence of severe pruritus and severe local skin pain were both significantly reduced in the Mebo group (14.1% in Biafine vs. 2.9% in Mebo, P=0.016 for pruritus and 11.5% vs. 1.4%, respectively, P=0.02 for severe pain). CONCLUSIONS: This study showed no difference between Mebo and Biafine in the incidence and severity of breast skin dermatitis during radiation therapy. However, the use of Mebo ointment was associated with decreased severe pruritus and pain which could positively affect patient comfort and quality of life.

Effectiveness of radial artery-based multiarterial coronary artery bypass grafting: Role of body habitus.

BACKGROUND: The multiarterial grafting survival advantage noted in the overall population undergoing coronary artery bypass grafting is not well defined in the obese. We investigated the early to late survival effects of the radial artery in left internal thoracic artery-based multiarterial bypass grafting (radial artery-multiarterial bypass grafting) versus single arterial bypass grafting (left internal thoracic artery-single arterial bypass grafting) in obese patients. METHODS: We analyzed 15-year Kaplan-Meier survival in 6102 patients receiving primary, left internal thoracic artery-based coronary artery bypass grafting with 2 or more grafts divided into body mass index groups: nonobese (<30 kg/m2) and all-obese, comprised of mildly obese (30-35 kg/m2) and morbidly obese (>35 kg/m2). Risk-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of radial artery-multiarterial bypass grafting versus left internal thoracic artery-single arterial bypass grafting were derived via Cox regression and applied separately for early (<0.5 years), intermediate (0.5-5 years), and late (5-15 years) follow-up in each body mass index cohort. Propensity score matching between radial artery-multiarterial bypass grafting and left internal thoracic artery-single arterial bypass grafting cohorts within the body mass index groups was performed as a corroborating analysis. RESULTS: Radial artery-multiarterial bypass grafting was more frequently used in obese patients who were younger (62 ± 10 years; mild/morbid: 45.4%/54.4% radial artery-multiarterial bypass grafting) compared with nonobese patients (66 ± 10 years; 37.4% radial artery-multiarterial bypass grafting). Unadjusted 15-year survival was significantly better for radial artery-multiarterial bypass grafting in all body mass index groups. Multivariate analysis showed a survival benefit of radial artery-multiarterial bypass grafting over the entire 0- to 15-year study period in the all-obese cohort (HR, 0.85; 95% CI, 0.74-0.98) and was more pronounced in the mildly obese (HR, 0.79; 95% CI, 0.66-0.96) versus morbidly obese (HR, 0.88; 95% CI, 0.69-1.13). The radial artery-multiarterial bypass grafting survival benefit was realized between 0.5 and 5 years postoperatively and was comparable for all-obese (HR, 0.69; 95% CI, 0.51-0.94) and nonobese (HR, 0.68; 95% CI, 0.52-0.88) groups. Propensity score matching was confirmatory. CONCLUSIONS: Radial artery-multiarterial bypass grafting confers a long-term survival advantage in both obese and nonobese patients.

Secondary Cleft Nasolabial Deformities: A New Classification System for Evaluation and Surgical Revision.

INTRODUCTION: Secondary or residual cleft lip and nasal deformities following primary unilateral or bilateral cleft lip repair are common. Many classification systems have been proposed to describe congenital cleft lip and palate deformities before repair. This article proposes a one-of-a-kind classification system for residual cleft deformities and describes its application to 136 cleft lip revision cases from cleft outreach missions worldwide. METHODS: Patients' demographics and deformities were classified preoperatively, and a database of the classification was created. Postoperatively, the type of surgery performed was added to the database and comparison was done using an independent t test. RESULTS: Kappa coefficient was 0.92 and showed excellent agreement between the type assigned preoperatively to the patient and the type of procedure done. CONCLUSIONS: This system proves to provide good descriptions of the deformities, is user friendly, facilitates the planning of the corrective surgical procedure, and enhances the communicative lingo between surgeons and members of cleft multidisciplinary care teams. It is broadly applicable in outreach missions with limited resources and cleft referral centers with considerable load.

Intraoperative Radiation Therapy: A Promising Treatment Modality in Head and Neck Cancer.

Every year, almost 62,000 are diagnosed with a head and neck cancer (HNC) and 13,000 will succumb to their disease. In the primary setting, intraoperative radiation therapy (IORT) can be used as a boost in select patients in order to optimize local control. Addition of external beam radiation to limited volumes results in improved disease control over surgery and IORT alone. In the recurrent setting, IORT can improve outcomes from salvage surgery especially in patients previously treated with external beam radiation. The use of IORT remains limited to select institutions with various modalities being currently employed including orthovoltage, electrons, and high-dose rate brachytherapy. Practically, execution of IORT requires a coordinated effort and careful planning by a multidisciplinary team involving the head and neck surgeon, radiation oncologist, and physicist. The current review summarizes common uses, outcomes, toxicities, and technical aspects of IORT in HNC patients.