Joseph Picone

Dr. Joseph Picone


Research Focus & Metrics

Academic Career Length

22Years (2005-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 December 2015
Verified
Conference Paper
Closed access
SLS. Lopez
S. Lopez
GSG. Suarez
G. Suarez
DJD. Jungreis
D. Jungreis
IOI. Obeid
I. Obeid
JPJoseph Picone
Joseph Picone
The interpretation of electroencephalograms (EEGs) is a process that is still dependent on the subjective analysis of the examiners. Though interrater agreement on critical events such as seizures is high, it is much lower on subtler events (e.g., when there are benign variants). The process used by an expert to interpret an EEG is quite subjective and hard to replicate by machine. The performance of machine learning technology is far from human performance. We have been developing an interpretation system, AutoEEG, with a goal of exceeding human performance on this task. In this work, we are focusing on one of the early decisions made in this process - whether an EEG is normal or abnormal. We explore two baseline classification algorithms: k-Nearest Neighbor (kNN) and Random Forest Ensemble Learning (RF). A subset of the TUH EEG Corpus was used to evaluate performance. Principal Components Analysis (PCA) was used to reduce the dimensionality of the data. kNN achieved a 41.8% detection error rate while RF achieved an error rate of 31.7%. These error rates are significantly lower than those obtained by random guessing based on priors (49.5%). The majority of the errors were related to misclassification of normal EEGs.

1 December 2015
Verified
Conference Paper
Open access
AHA. Harati
A. Harati
MGM. Golmohammadi
M. Golmohammadi
SLS. Lopez
S. Lopez
IOI. Obeid
I. Obeid
JPJoseph Picone
Joseph Picone
Feature extraction for automatic classification of EEG signals typically relies on time frequency representations of the signal. Techniques such as cepstral-based filter banks or wavelets are popular analysis techniques in many signal processing applications including EEG classification. In this paper, we present a comparison of a variety of approaches to estimating and postprocessing features. To further aid in discrimination of periodic signals from aperiodic signals, we add a differential energy term. We evaluate our approaches on the TUH EEG Corpus, which is the largest publicly available EEG corpus and an exceedingly challenging task due to the clinical nature of the data. We demonstrate that a variant of a standard filter bank-based approach, coupled with first and second derivatives, provides a substantial reduction in the overall error rate. The combination of differential energy and derivatives produces a 24% absolute reduction in the error rate and improves our ability to discriminate between signal events and background noise. This relatively simple approach proves to be comparable to other popular feature extraction approaches such as wavelets, but is much more computationally efficient.

1 December 2016
Verified
Conference Paper
Closed access
SLS. Lopez
S. Lopez
AGA. Gross
A. Gross
SYS. Yang
S. Yang
MGM. Golmohammadi
M. Golmohammadi
IOI. Obeid
I. Obeid
JPJoseph Picone
Joseph Picone
Clinical electroencephalographic (EEG) data varies significantly depending on a number of operational conditions (e.g., the type and placement of electrodes, the type of electrical grounding used). This investigation explores the statistical differences present in two different referential montages: Linked Ear (LE) and Averaged Reference (AR). Each of these accounts for approximately 45% of the data in the TUH EEG Corpus. In this study, we explore the impact this variability has on machine learning performance. We compare the statistical properties of features generated using these two montages, and explore the impact of performance on our standard Hidden Markov Model (HMM) based classification system. We show that a system trained on LE data significantly outperforms one trained only on AR data (77.2% vs. 61.4%). We also demonstrate that performance of a system trained on both data sets is somewhat compromised (71.4% vs. 77.2%). A statistical analysis of the data suggests that mean, variance and channel normalization should be considered. However, cepstral mean subtraction failed to produce an improvement in performance, suggesting that the impact of these statistical differences is subtler.


23 April 2020 in IEEE Transactions on Biomedical Engineering
Verified
Journal Article
Closed access
SISayemul Islam
Sayemul Islam
VSVinit Shah
Vinit Shah
SGSai Teja Reddy Gidde
Sai Teja Reddy Gidde
PHParsaoran Hutapea
Parsaoran Hutapea
SSSeung Hyun Song
Seung Hyun Song
Albert KimAlbert Kim
Albert Kim
A leading cause of traumatic brain injury (TBI) is intracranial brain deformation due to mechanical impact. This deformation is viscoelastic and differs from a traditional rigid transformation. In this paper, we describe a machine learning enabled wireless sensing system that predicts the trajectory of intracranial brain deformation. The sensing system consists of an implantable soft magnet and an external magnetic sensor array with a sensing volume of 12 × 12 × 4 mm3. Machine learning algorithm predicts the brain deformation by interpreting the magnetic sensor outputs created by the change in position of the implanted soft magnet. Three different machine learning models were trained on calibration data: (1) random forests, (2) k-nearest neighbors, and (3) a multi-layer perceptron-based neural network. These models were validated using both in vitro (a needle inserted into PVC gel) and in vivo (blast exposure to live and dead rat brains) experiments. The in vitro gel deformation predicted by these machine learning models showed excellent agreement with the camera measurements, and had absolute error = 138 μm, Fréchet distance = 372 μm with normalized Procrustes disparity = 0.034. The in vivo brain deformation predicted by these models had absolute error = 50 μm, Fréchet distance = 95 μm with normalized Procrustes disparity = 0.055 for dead animal and absolute error = 125 μm, Fréchet distance = 289 μm with normalized Procrustes disparity = 0.2 for live animal respectively. These results suggest that the proposed machine learning enabled sensor system can be an effective tool for measuring in situ brain deformation.

24 March 2005
Verified
Conference Paper
Closed access
JPJoseph Picone
Joseph Picone
GDG. Doddington
G. Doddington