Unified detection and tracking of instruments during retinal microsurgery

Raphael Sznitman, Rogerio Richa, Russell H Taylor, Bruno Jedynak, Gregory Hager

Research output: Contribution to journalArticle

Abstract

Methods for tracking an object have generally fallen into two groups: tracking by detection and tracking through local optimization. The advantage of detection-based tracking is its ability to deal with target appearance and disappearance, but it does not naturally take advantage of target motion continuity during detection. The advantage of local optimization is efficiency and accuracy, but it requires additional algorithms to initialize tracking when the target is lost. To bridge these two approaches, we propose a framework for unified detection and tracking as a time-series Bayesian estimation problem. The basis of our approach is to treat both detection and tracking as a sequential entropy minimization problem, where the goal is to determine the parameters describing a target in each frame. To do this we integrate the Active Testing (AT) paradigm with Bayesian filtering, and this results in a framework capable of both detecting and tracking robustly in situations where the target object enters and leaves the field of view regularly. We demonstrate our approach on a retinal tool tracking problem and show through extensive experiments that our method provides an efficient and robust tracking solution.

Original languageEnglish (US)
Article number6319313
Pages (from-to)1263-1273
Number of pages11
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume35
Issue number5
DOIs
StatePublished - 2013

Fingerprint

Target
Time series
Entropy
Local Optimization
Testing
Experiments
Bayesian Estimation
Field of View
Minimization Problem
Filtering
Paradigm
Integrate
Motion
Demonstrate
Experiment
Framework
Object

Keywords

  • active testing
  • adaptive sensing
  • instrument tracking
  • retinal microsurgery
  • Unified object detection and tracking

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Software
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

Unified detection and tracking of instruments during retinal microsurgery. / Sznitman, Raphael; Richa, Rogerio; Taylor, Russell H; Jedynak, Bruno; Hager, Gregory.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, No. 5, 6319313, 2013, p. 1263-1273.

Research output: Contribution to journalArticle

@article{c342361a832f4712b03cc6e2e70c16d4,
title = "Unified detection and tracking of instruments during retinal microsurgery",
abstract = "Methods for tracking an object have generally fallen into two groups: tracking by detection and tracking through local optimization. The advantage of detection-based tracking is its ability to deal with target appearance and disappearance, but it does not naturally take advantage of target motion continuity during detection. The advantage of local optimization is efficiency and accuracy, but it requires additional algorithms to initialize tracking when the target is lost. To bridge these two approaches, we propose a framework for unified detection and tracking as a time-series Bayesian estimation problem. The basis of our approach is to treat both detection and tracking as a sequential entropy minimization problem, where the goal is to determine the parameters describing a target in each frame. To do this we integrate the Active Testing (AT) paradigm with Bayesian filtering, and this results in a framework capable of both detecting and tracking robustly in situations where the target object enters and leaves the field of view regularly. We demonstrate our approach on a retinal tool tracking problem and show through extensive experiments that our method provides an efficient and robust tracking solution.",
keywords = "active testing, adaptive sensing, instrument tracking, retinal microsurgery, Unified object detection and tracking",
author = "Raphael Sznitman and Rogerio Richa and Taylor, {Russell H} and Bruno Jedynak and Gregory Hager",
year = "2013",
doi = "10.1109/TPAMI.2012.209",
language = "English (US)",
volume = "35",
pages = "1263--1273",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "5",

}

TY - JOUR

T1 - Unified detection and tracking of instruments during retinal microsurgery

AU - Sznitman, Raphael

AU - Richa, Rogerio

AU - Taylor, Russell H

AU - Jedynak, Bruno

AU - Hager, Gregory

PY - 2013

Y1 - 2013

N2 - Methods for tracking an object have generally fallen into two groups: tracking by detection and tracking through local optimization. The advantage of detection-based tracking is its ability to deal with target appearance and disappearance, but it does not naturally take advantage of target motion continuity during detection. The advantage of local optimization is efficiency and accuracy, but it requires additional algorithms to initialize tracking when the target is lost. To bridge these two approaches, we propose a framework for unified detection and tracking as a time-series Bayesian estimation problem. The basis of our approach is to treat both detection and tracking as a sequential entropy minimization problem, where the goal is to determine the parameters describing a target in each frame. To do this we integrate the Active Testing (AT) paradigm with Bayesian filtering, and this results in a framework capable of both detecting and tracking robustly in situations where the target object enters and leaves the field of view regularly. We demonstrate our approach on a retinal tool tracking problem and show through extensive experiments that our method provides an efficient and robust tracking solution.

AB - Methods for tracking an object have generally fallen into two groups: tracking by detection and tracking through local optimization. The advantage of detection-based tracking is its ability to deal with target appearance and disappearance, but it does not naturally take advantage of target motion continuity during detection. The advantage of local optimization is efficiency and accuracy, but it requires additional algorithms to initialize tracking when the target is lost. To bridge these two approaches, we propose a framework for unified detection and tracking as a time-series Bayesian estimation problem. The basis of our approach is to treat both detection and tracking as a sequential entropy minimization problem, where the goal is to determine the parameters describing a target in each frame. To do this we integrate the Active Testing (AT) paradigm with Bayesian filtering, and this results in a framework capable of both detecting and tracking robustly in situations where the target object enters and leaves the field of view regularly. We demonstrate our approach on a retinal tool tracking problem and show through extensive experiments that our method provides an efficient and robust tracking solution.

KW - active testing

KW - adaptive sensing

KW - instrument tracking

KW - retinal microsurgery

KW - Unified object detection and tracking

UR - http://www.scopus.com/inward/record.url?scp=84875465090&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875465090&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2012.209

DO - 10.1109/TPAMI.2012.209

M3 - Article

C2 - 23520263

AN - SCOPUS:84875465090

VL - 35

SP - 1263

EP - 1273

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 5

M1 - 6319313

ER -