Jonathan Stoeckel

Multiple-Instance Learning Improves CAD Detection of Masses in Digital Mammography

We propose a novel multiple-instance learning(MIL) algorithm for designing classifiers for use in computer aided detection(CAD). The proposed algorithm has 3 advantages over classical methods. First, unlike traditional learning algorithms that minimize the candidate level misclassification error, the proposed algorithm directly optimizes the patient-wise sensitivity. Second, this algorithm automatically selects a small subset of statistically useful features. Third, this algorithm is very fast, utilizes all of the available training data (without the need for cross-validation etc.), and requires no human hand tuning or intervention. Experimentally the algorithm is more accurate than state of the art support vector machine (SVM) classifier, and substantially reduces the number of features that have to be computed.

A novel method of partitioning regions in lungs and their usage in feature extraction for reducing false positives

Chest X-ray (CXR) data is a 2D projection image. The main drawback of such an image is that each pixel of it represents a volumetric integration. This poses a challenge in detection and estimation of nodules and their characteristics. Due to human anatomy there are a lot of lung structures which can be falsely identified as nodules in a projection data. Detection of nodules with a large number of false positives (FP) adds more work for the radiologists. With the help of CAD algorithms we aim to identify regions which cause higher FP readings or provide additional information for nodule detection based on the human anatomy. Different lung regions have different image characteristics we take advantage of this and propose an automatic lung partitioning into vessel, apical, basal and exterior pulmonary regions. Anatomical landmarks like aortic arch and end of cardiac-notch along-with inter intra-rib width and their shape characteristics were used for this partitioning. Likelihood of FPs is more in vessel, apical and exterior pulmonary regions due to rib-crossing, overlap of vessel with rib and vessel branching. For each of these three cases, special features were designed based on histogram of rib slope and the structural properties of rib segments information. These features were assigned different weights based on the partitioning. An experiment was carried out using a prototype CAD system 150 routine CXR studies were acquired from three institutions (24 negatives, rest with one or more nodules). Our algorithm provided a sensitivity of 70.4% with 5 FP/image for cross-validation without partition. Inclusion of the proposed techniques increases the sensitivity to 78.1% with 4.1 FP/image.

Use of random process-based fractal measure for characterization nodules and suspicious regions in lung

Chest X-ray (CXR) data is a projection image where each pixel of it represents a volumetric integration. Consequently identification of nodules and their characteristics is a difficult task in such images. Using a novel application of random process-based fractal image processing technique we extract features for nodule characterization. The uniqueness of the proposed technique lies in the fact that instead of relying on apriori information from user as in other random process inspired measures, we translate the random walk process into a feature which is based on its realization values. The Normalized Fractional Brownian Motion (NFBM) Model is derived from the random walk process. Using neighborhood region information in an incremental manner we can characterize the smoothness or roughness of a surface. The NFBM system gives a measure of roughness of a surface which in our case is a suspicious region (probable nodule). A classification procedure uses this measure to categorize nodule and non-nodule structures in the lung. The NFBM feature set is integrated in a prototype CAD system for nodule detection in CXR. Our algorithm provided a sensitivity of 75.9% with 3.1 FP/image on an independent test set of 50 CXR studies.

Addressing image variability while learning classifiers for detecting clusters of micro-calcifications

Computer aided detection systems for mammography typically use standard classification algorithms from machine learning for detecting lesions. However, these general purpose learning algorithms make implicit assumptions that are commonly violated in CAD problems. We propose a new ensemble algorithm that explicitly accounts for the small fraction of outlier images which tend to produce a large number of false positives. A bootstrapping procedure is used to ensure that the candidates from these outlier images do not skew the statistical properties of the training samples. Experimental studies on the detection of clusters of micro-calcifications indicate that the proposed method significantly outperforms a state-of-the-art general purpose method for designing classifiers (SVM), in terms of FROC curves on a hold out test set.

How can image quality affect the detection performance of breast CAD (computer aided detection) in FFDM (full field digital mammography)? – a comparative study with two different FFDM systems-

This study was conducted to retrospectively evaluate the variation of CAD performance utilizing two different FFDM systems in normal clinical cases.

A fast algorithm for body extraction in CT volumes

The Computed Tomography (CT) modality shows not only the body of the patient in the volumes it generates, but also the clothing, the cushion and the table. This might be a problem especially for two applications. The first is 3D visualization, where the table has high density parts that might hide regions of interest. The second is registration of acquisitions obtained at different time points; indeed, the table and cushions might be visible in one data set only, and their positions and shapes may vary, making the registration less accurate. An automatic approach for extracting the body would solve those problems. It should be robust, reliable, and fast. We therefore propose a multi-scale method based on deformable models. The idea is to move a surface across the image that attaches to the boundaries of the body. We iteratively compute forces which take into account local information around the surface. Those make it move through the table but ensure that it stops when coming close to the body. Our model has elastic properties; moreover, we take into account the fact that some regions in the volume convey more information than others by giving them more weight. This is done by using normalized convolution when regularizing the surface. The algorithm*, tested on a database of over a hundred volumes of whole body, chest or lower abdomen, has proven to be very efficient, even for volumes with up to 900 slices, providing accurate results in an average time of 6 seconds. It is also robust against noise and variations of scale and tables shape.

Towards learning spiculation score of the masses in mammography images

This paper deals with learning spiculation scores of masses in a supervised manner Three spiculation score prediction models treating the score either as a continuous or ordinary variable are presented These models were compared on a data-set of 255 masses.

An approach to automatic detection of body parts and their size estimation from computed tomography image

Computer-aided diagnosis (CAD) systems usually require information about regions of interest in images, like: lungs (for nodule detection), colon (for identifying polyps), etc. Many times, it is computationally intensive to process large data sets as in CT to find these areas of interest. A fast and accurate recognition of the different regions of interest in the human body from images is therefore necessary. In this paper we propose a fast and efficient algorithm that can detect the organs of interest in a CT volume and estimate their sizes. Instead of analyzing the whole 3D volume; which is computationally expensive, a binary search technique is adapted to search in a few slices. The slices selected in the search process is segmented and different regions are labeled. Decision over whether the image belongs to a lung or colon or both is based on the count of lung/colon pixels in the slice. Once the detection is done we look for the start and end slice of the body part to have an estimate of their sizes. The algorithm involves certain search decisions based on some predefined threshold values which are empirically chosen from a training data set. The effectiveness of our technique is confirmed by applying it on an independent test data set. Detection accuracy of 100% is obtained on a test set. This algorithm can be integrated in a CAD system for running the right application, or can be used in pre-sets for visualization purposes and other post-processing like image registration etc.