Intracerebral Hemorrhage Segmentation on Noncontrast Computed Tomography Using a Masked Loss Function U-Net Approach.

OBJECTIVE: Intracerebral hemorrhage (ICH) volume is a strong predictor of outcome in patients presenting with acute hemorrhagic stroke. It is necessary to segment the hematoma for ICH volume estimation and for computerized extraction of features, such as spot sign, texture parameters, or extravasated iodine content at dual-energy computed tomography. Manual and semiautomatic segmentation methods to delineate the hematoma are tedious, user dependent, and require trained personnel. This article presents a convolutional neural network to automatically delineate ICH from noncontrast computed tomography scans of the head. METHODS: A model combining a U-Net architecture with a masked loss function was trained on standard noncontrast computed tomography images that were down sampled to 256 × 256 size. Data augmentation was applied to prevent overfitting, and the loss score was calculated using the soft Dice loss function. The Dice coefficient and the Hausdorff distance were computed to quantitatively evaluate the segmentation performance of the model, together with the sensitivity and specificity to determine the ICH detection accuracy. RESULTS: The results demonstrate a median Dice coefficient of 75.9% and Hausdorff distance of 2.65 pixels in segmentation performance, with a detection sensitivity of 77.0% and specificity of 96.2%. CONCLUSIONS: The proposed masked loss U-Net is accurate in the automatic segmentation of ICH. Future research should focus on increasing the detection sensitivity of the model and comparing its performance with other model architectures.

Photographic documentation and severity quantification of pectus excavatum through three-dimensional optical surface imaging.

Conventional photography is commonly used to visually document pectus excavatum and objectively assess chest wall changes over time without repeated exposure to ionising radiation, as in our centre since 2008. However, as conventional photography is labour-intensive and lacks three-dimensional (3D) information that is essential in 3D deformities like pectus excavatum, we developed a novel imaging and processing protocol based on 3D optical surface imaging. The objective of this study was to report our developed protocol to visually document pectus excavatum through 3D imaging. We also investigated the absolute agreement of the 3D image- and conventional photography-derived pectus excavatum depth to investigate whether both techniques could be used interchangeably to measure pectus excavatum depth and assess its evolution. The protocol consisted of three consecutive steps: patient positioning and instructions, data acquisition, and data processing. Three-dimensional imaging through the developed protocol was feasible for all 19 participants. The 3D image- and photography-derived pectus excavatum depth demonstrated good to excellent agreement (intraclass correlation coefficient: 0.97; 95%-confidence interval: 0.88 to 0.99; p < 0.001). In conclusion, 3D imaging through the developed protocol is a feasible and attractive alternative to document the surface geometry of pectus excavatum and can be used interchangeably with conventional photography to determine pectus severity. Clinical registration number: NCT04185870.

Pectus excavatum and carinatum: a narrative review of epidemiology, etiopathogenesis, clinical features, and classification.

Background and Objective: A wide variety of congenital chest wall deformities that manifest in infants, children and adolescents exists, among which are pectus excavatum and pectus carinatum. Numerous studies have been conducted over the years aiming to better understand these deformities. This report provides a brief overview of what is currently known about the epidemiology, etiopathogenesis, clinical presentation, and classification of these deformities, and highlights the gaps in knowledge. Methods: A search was conducted for all the above-described domains in the PubMed and Embase databases. Key Content and Findings: A total of 147 articles were included in this narrative review. Estimation of the true incidence and prevalence of pectus excavatum and carinatum is challenging due to lacking consensus on a definition of both deformities. Nowadays, several theories for the development of pectus excavatum and carinatum have been suggested which focus on intrinsic or extrinsic pathogenic factors, with the leading hypothesis focusing on overgrowth or growth disturbance of costal cartilages. Furthermore, genetic predisposition to the deformities is likely to exist. Pectus excavatum is frequently associated with cardiopulmonary symptoms, while pectus carinatum patients mostly present with cosmetic complaints. Both deformities are classified based on the shape or severity of the deformity. However, each classification system has its limitations. Conclusions: Substantial progress has been made in the past few decades in understanding the development and symptomatology of pectus excavatum and carinatum. Current hypotheses on the etiology of the deformities should be confirmed by biomedical and genetic studies. For clinical purposes, the establishment of a clear definition and classification system for both deformities based on objective morphologic features is eagerly anticipated.

Advancements in preoperative imaging of pectus excavatum: a comprehensive review.

Pectus excavatum, the most common pectus deformity, varies in severity and has been associated with cardiopulmonary impairment and psychological distress. Since its initial documentation, a multitude of imaging techniques for preoperative evaluation (i.e., diagnosis, severity classification, functional assessment, and surgical planning) have been reported. Conventional imaging techniques encompass computed tomography (CT), chest radiography, magnetic resonance imaging (MRI), echocardiography and medical photography, while three dimensional (3D) optical surface imaging is a promising emerging technique in the preoperative assessment of pectus excavatum. This narrative review explores the current insights and advancements of these imaging modalities. CT imaging allows for the calculation of pectus indices and evaluation of cardiac compression and displacement. Recent developments focus on automated calculations, minimizing radiation exposure and improving surgical planning. Chest radiography offers a radiation-reducing alternative for pectus index measurement, but is unsuitable for disproportionally asymmetric chest deformations. MRI is a radiation-free imaging method, and allows for the calculation of pectus indices as well as the assessment of cardiac function. Real-time MRI provides dynamic insights, while exercise MRI shows promise for comprehensive evaluation of cardiac function but requires additional developments. Using echocardiography, structural cardiac changes can be identified, but its use in evaluating cardiac function in pectus excavatum patients is limited. Medical photography combined with caliper measurements complements other imaging methods for qualitative and quantitative documentation of pectus excavatum. Emerging as an innovative technique, 3D optical surface imaging offers a rapid, radiation-free assessment of the deformity which correlates with conventional pectus indices. Potential applications include quantifying other morphological features and predicting cardiac compression. However, standardization and validation are needed for its widespread use. This review provides an overview of preoperative imaging of pectus excavatum, highlighting the current developments in conventional methods and the potential of the emerging 3D optical surface imaging technique. These advancements hold promise for the future of the assessment and surgical planning of pectus excavatum.

Predicting Aesthetic Outcome of the Nuss Procedure in Patients with Pectus Excavatum.

Patients suffering from pectus excavatum often experience psychosocial distress due to perceived anomalies in their physical appearance. The ability to visually inform patients about their expected aesthetic outcome after surgical correction is still lacking. This study aims to develop an automatic, patient-specific model to predict aesthetic outcome after the Nuss procedure. Patients prospectively received preoperative and postoperative 3-dimensional optical surface scanning of their chest during the Nuss procedure. A prediction model was composed based on nonlinear least squares data-fitting, regression methods and a 2-dimensional Gaussian function with adjustable amplitude, variance, rotation, skewness, and kurtosis components. Morphological features of pectus excavatum were extracted from preoperative images using a previously developed surface analysis tool to generate a patient-specific model. Prediction accuracy was evaluated through cross-validation, utilizing the mean root squared deviation and maximum positive and negative deviations as performance measures. The prediction model was evaluated on 30 (90% male) prospectively imaged patients. The model achieved an average root mean squared deviation of 6.3 ± 2.0 mm, with average maximum positive and negative deviations of 12.7 ± 6.1 and -10.2 ± 5.7 mm, respectively, between the predicted and actual postoperative aesthetic result. Our developed 2-dimensional Gaussian model based on 3-dimensional optical surface images is a clinically promising tool to predict postsurgical aesthetic outcome in patients with pectus excavatum. Prediction of the aesthetic outcome after the Nuss procedure potentially improves information provision and expectation management among patients. Further research should assess whether increasing the sample size may reduce deviations and improve performance.

Pectus Excavatum: Consensus and Controversies in Clinical Practice.

BACKGROUND: Pectus excavatum is the most common congenital anterior chest wall deformity. Currently, a wide variety of diagnostic protocols and criteria for corrective surgery are being used. Their use is predominantly based on local preferences and experience. To date, no guideline is available, introducing heterogeneity of care as observed in current daily practice. The aim of this study was to evaluate consensus and controversies regarding the diagnostic protocol, indications for surgical correction, and postoperative evaluation of pectus excavatum. METHODS: The study consisted of 3 consecutive survey rounds evaluating agreement on different statements regarding pectus excavatum care. Consensus was achieved if at least 70% of participants provided a concurring opinion. RESULTS: All 3 rounds were completed by 57 participants (18% response rate). Consensus was achieved on 18 of 62 statements (29%). Regarding the diagnostic protocol, participants agreed to routinely include conventional photography. In the presence of cardiac impairment, electrocardiography and echocardiography were indicated. Upon suspicion of pulmonary impairment, spirometry was recommended. In addition, consensus was reached on the indications for corrective surgery, including symptomatic pectus excavatum and progression. Participants moreover agreed that a plain chest radiograph must be acquired directly after surgery, whereas conventional photography and physical examination should both be part of routine postoperative follow-up. CONCLUSIONS: Through a multiround survey, international consensus was formed on multiple topics to aid standardization of pectus excavatum care.

Three-dimensional Surface Imaging for Clinical Decision Making in Pectus Excavatum.

To evaluate pectus excavatum, 3-dimensional surface imaging is a promising radiation-free alternative to computed tomography and plain radiographs. Given that 3-dimensional images concern the external surface, the conventional Haller index, and correction index are not applicable as these are based on internal diameters. Therefore, external equivalents have been introduced for 3-dimensional images. However, cut-off values to help determine surgical candidacy using external indices are lacking. A prospective cohort study was conducted. Consecutive patients referred for suspected pectus excavatum received a computed tomography (≥18 years) or plain radiographs (<18 years). The external Haller index and external correction index were calculated from additionally acquired 3-dimensional images. Cut-off values for the 3-dimensional image derived indices were obtained by receiver-operating characteristic curve analyses, using a conventional Haller index ≥3.25, and computed tomography derived correction index ≥28.0% as indicative for surgery. Sixty-one and 63 patients were included in the computed tomography and radiograph group, respectively. To determine potential surgical candidacy, receiver-operating characteristic analyses found an optimum cut-off of ≥1.83 for the external Haller index in both the computed tomography and radiograph group with a positive predictive value between 0.90 and 0.97 and a negative predictive value between 0.72 and 0.81. The optimal cut-off for the external correction index was ≥15.2% with a positive predictive value of 0.86 and negative predictive value of 0.93. The 3-dimensional image derived external Haller index and external correction index are accurate radiation-free alternatives to facilitate surgical decision-making among patients suspected of pectus excavatum with values of ≥1.83 and ≥15.2% indicative for surgery.

The Automatic Quantification of Morphological Features of Pectus Excavatum Based on Three-Dimensional Images.

Visual examination and quantification of severity are essential for clinical decision making in patients with pectus excavatum. Yet, visual assessment is prone to inter- and intra-observer variability and current quantitative methods are inadequate. This study aims to develop and evaluate a novel, automatic and non-invasive method to objectively quantify pectus excavatum morphology based on three-dimensional images. Key steps of the automatic analysis are normalization of image orientation, slicing, and computation of the morphological features encompassing pectus depth, width, length, volume, position, steepness, flaring, asymmetry and mean cross-sectional area. A digital phantom mimicking a patient with pectus excavatum was used to verify the analysis method. Prospective three-dimensional imaging and subsequent surface analysis in patients with pectus excavatum was performed to assess clinical feasibility. Verification of the developed analysis tool demonstrated 100% reproducibility of all morphological feature values. Calculated parameters compared to the predetermined phantom dimensions were accurate for all but four features. The pectus width, length, volume and steepness showed an error of 4 mm (4%), 2 mm (2%), 12 mL (5%) and 1 degree (3%), respectively. Prospective imaging of 52 patients (88% males) demonstrated the feasibility of the developed tool to quantify morphological features of pectus excavatum in the clinical setting. Mean duration to calculate all features in one patient was 7.6 seconds. We have developed and presented a non-invasive pectus excavatum surface analysis tool, that is feasible to automatically quantify morphological features based on three-dimensional images with promising accuracy and reproducibility.

Sternal elevation by the crane technique during pectus excavatum repair: A quantitative analysis.

INTRODUCTION: The crane technique is used to facilitate sternal elevation to provide safe mediastinal passage during the Nuss procedure. The aim was to objectively quantitate the elevation of the crane by 3-dimensional chest images acquired during the Nuss procedure. METHODS: A prospective cohort study was conducted. Patients undergoing the Nuss procedure were eligible. Sternal elevation was achieved by the crane technique providing a simultaneous lift of the anterior chest wall and reduction of the pectus excavatum depth. Both effects were evaluated. Three-dimensional surface images were acquired before incision, following sternal lift, and after bar implantation and quantitatively compared. Reduction of the external pectus excavatum depth was expressed as a percentage. RESULTS: Thirty patients were included. Ninety percent were male, with a median age of 15.5 years (interquartile range [IQR], 14.5-17.4), Haller index of 3.56 (IQR, 3.09-4.65), and external pectus depth of 18 mm (IQR, 11-23). Sternal elevation by the crane provided a median 78% (IQR, 63-100) reduction of the deformity, corresponding with a residual depth of 3 mm (IQR, 0-7). The percentual reduction diminished with increasing depth of the sternal depression (correlation, -0.86). Besides reducing the deformity, the crane caused an elevation of the anterior chest over a large surface area with a maximum lift of 26 mm (IQR, 19-32). CONCLUSIONS: The crane is an effective sternal elevation technique, providing 78% reduction of the sternal depression, although its effect lessens with increasing depth. In addition, it produces an elevation of the anterior chest over a large surface area.