Structural Analysis and Optimization of Convolutional Neural Networks with a Small Sample Size.

Deep neural networks have gained immense popularity in the Big Data problem; however, the availability of training samples can be relatively limited in specific application domains, particularly medical imaging, and consequently leading to overfitting problems. This "Small Data" challenge may need a mindset that is entirely different from the existing Big Data paradigm. Here, under the small data scenarios, we examined whether the network structure has a substantial influence on the performance and whether the optimal structure is predominantly determined by sample size or data nature. To this end, we listed all possible combinations of layers given an upper bound of the VC-dimension to study how structural hyperparameters affected the performance. Our results showed that structural optimization improved accuracy by 27.99%, 16.44%, and 13.11% over random selection for a sample size of 100, 500, and 1,000 in the MNIST dataset, respectively, suggesting that the importance of the network structure increases as the sample size becomes smaller. Furthermore, the optimal network structure was mostly determined by the data nature (photographic, calligraphic, or medical images), and less affected by the sample size, suggesting that the optimal network structure is data-driven, not sample size driven. After network structure optimization, the convolutional neural network could achieve 91.13% accuracy with only 500 samples, 93.66% accuracy with only 1000 samples for the MNIST dataset and 94.10% accuracy with only 3300 samples for the Mitosis (microscopic) dataset. These results indicate the primary importance of the network structure and the nature of the data in facing the Small Data challenge.

Machine Learning Versus Logistic Regression Methods for 2-Year Mortality Prognostication in a Small, Heterogeneous Glioma Database.

BACKGROUND: Machine learning (ML) is the application of specialized algorithms to datasets for trend delineation, categorization, or prediction. ML techniques have been traditionally applied to large, highly dimensional databases. Gliomas are a heterogeneous group of primary brain tumors, traditionally graded using histopathologic features. Recently, the World Health Organization proposed a novel grading system for gliomas incorporating molecular characteristics. We aimed to study whether ML could achieve accurate prognostication of 2-year mortality in a small, highly dimensional database of patients with glioma. METHODS: We applied 3 ML techniques (artificial neural networks [ANNs], decision trees [DTs], and support vector machines [SVMs]) and classical logistic regression (LR) to a dataset consisting of 76 patients with glioma of all grades. We compared the effect of applying the algorithms to the raw database versus a database where only statistically significant features were included into the algorithmic inputs (feature selection). RESULTS: Raw input consisted of 21 variables and achieved performance of accuracy/area (C.I.) under the curve of 70.7%/0.70 (49.9-88.5) for ANN, 68%/0.72 (53.4-90.4) for SVM, 66.7%/0.64 (43.6-85.0) for LR, and 65%/0.70 (51.6-89.5) for DT. Feature selected input consisted of 14 variables and achieved performance of 73.4%/0.75 (62.9-87.9) for ANN, 73.3%/0.74 (62.1-87.4) for SVM, 69.3%/0.73 (60.0-85.8) for LR, and 65.2%/0.63 (49.1-76.9) for DT. CONCLUSIONS: We demonstrate that these techniques can also be applied to small, highly dimensional datasets. Our ML techniques achieved reasonable performance compared with similar studies in the literature. Although local databases may be small versus larger cancer repositories, we demonstrate that ML techniques can still be applied to their analysis; however, traditional statistical methods are of similar benefit.

Feasibility of Clinician-Facilitated Three-Dimensional Printing of Synthetic Cranioplasty Flaps.

BACKGROUND: Integration of three-dimensional (3D) printing and stereolithography into clinical practice is in its nascence, and concepts may be esoteric to the practicing neurosurgeon. Currently, creation of 3D printed implants involves recruitment of offsite third parties. We explored a range of 3D scanning and stereolithographic techniques to create patient-specific synthetic implants using an onsite, clinician-facilitated approach. METHODS: We simulated bilateral craniectomies in a single cadaveric specimen. We devised 3 methods of creating stereolithographically viable virtual models from removed bone. First, we used preoperative and postoperative computed tomography scanner-derived bony window models from which the flap was extracted. Second, we used an entry-level 3D light scanner to scan and render models of the individual bone pieces. Third, we used an arm-mounted, 3D laser scanner to create virtual models using a real-time approach. RESULTS: Flaps were printed from the computed tomography scanner and laser scanner models only in a ultraviolet-cured polymer. The light scanner did not produce suitable virtual models for printing. The computed tomography scanner-derived models required extensive postfabrication modification to fit the existing defects. The laser scanner models assumed good fit within the defects without any modification. CONCLUSIONS: The methods presented varying levels of complexity in acquisition and model rendering. Each technique required hardware at varying in price points from $0 to approximately $100,000. The laser scanner models produced the best quality parts, which had near-perfect fit with the original defects. Potential neurosurgical applications of this technology are discussed.