About

Siamak Ardekani is a professor affiliated with Johns Hopkins Biomedical Engineering and the Department of Medicine. With a background in medicine and medical imaging, he focuses on developing computational models to extract clinically relevant information from medical images generated through magnetic resonance imaging and computed tomography. His work involves creating image-based statistical shape and function models for early disease detection, monitoring response to therapy, and validating new therapeutic strategies in neurological and cardiovascular diseases. Ardekani's models are designed to assist physicians in identifying patients at high risk of complications and in monitoring their responsiveness to treatment. He earned his PhD in biomedical engineering from the University of California, Los Angeles, an MD from Shiraz University of Medical Sciences, and an MS in biomedical engineering from Drexel University. He has authored more than 20 research journals and conference papers in international journals and conference proceedings. His research aims to improve diagnostic and therapeutic strategies through advanced computational modeling of medical images, contributing to the fields of neuroimaging and cardiovascular imaging.

Research topics

• Computer Science
• Artificial Intelligence
• Computer vision
• Medicine
• Mathematics
• Algorithm

Selected publications

• Cardiac MRI Tagline Extraction Based on Diffeomorphic Active Contour Algorithm

  Lecture notes in computer science · 2023

  Senior authorCorresponding
  • Computer Science
  • Artificial Intelligence
  • Computer Science
  PublisherDOI

• 3D Attention M-net for Short-axis Left Ventricular Myocardium Segmentation in Mice MR cardiac Images

  2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) · 2021 · 1 citations

  Senior authorCorresponding
  • Artificial Intelligence
  • Computer Science
  • Artificial Intelligence

  Small rodent cardiac magnetic resonance imaging (MRI) plays an important role in preclinical models of cardiac disease. Accurate myocardial boundaries delineation is crucial to most morphological and functional analysis in rodent cardiac MRIs. However, rodent cardiac MRIs, due to animal's small cardiac volume and high heart rate, are usually acquired with sub-optimal resolution and low signal-to-noise ratio (SNR). These rodent cardiac MRIs can also suffer from signal loss due to the intra-voxel dephasing. These factors make automatic myocardial segmentation challenging. Manual contouring could be applied to label myocardial boundaries but it is usually laborious, time consuming, and not systematically objective. In this study, we present a deep learning approach based on 3D attention M-net to perform automatic segmentation of left ventricular myocardium. In the deep learning architecture, we use dual spatial-channel attention gates between encoder and decoder along with multi-scale feature fusion path after decoder. Attention gates enable networks to focus on relevant spatial information and channel features to improve segmentation performance. A distance derived loss term, besides general dice loss and binary cross entropy loss, was also introduced to our hybrid loss functions to refine segmentation contours. The proposed model outperforms other generic models, like U-Net and FCN, in major segmentation metrics including the dice score (0.9072), Jaccard index (0.8307) and Hausdorff distance (3.1754 pixels), which are comparable to the results achieved by state-of-the-art models on human cardiac ACDC17 datasets.Clinical relevance Small rodent cardiac MRI is routinely used to probe the effect of individual genes or groups of genes on the etiology of a large number of cardiovascular diseases. An automatic myocardium segmentation algorithm specifically designed for these data can enhance accuracy and reproducibility of cardiac structure and function analysis.

  PublisherOA PDFDOI

• Diffeomorphic Upsampling of Serially Acquired Sparse 2D Cross-Sections in Cardiac MRI

  2019-07-01 · 3 citations

  articleOpen accessSenior author

  In this paper, we propose a new technique for interpolating shapes in order to upsample a sparsely acquired serial-section image stack. The method is based on a maximum a posteriori estimation strategy which models neighboring sections as observations of random deformations of an image to be estimated. We show the computation of diffeomorphic trajectories between observed sections and define estimated upsampled image sections as a Jacobian-weighted sum of flowing images at corresponding distances along those trajectories. We apply this methodology to upsample stacks of sparse 2D magnetic resonance cross-sections through live mouse hearts. We show that the proposed method results in smoother and more accurate reconstructions over linear interpolation, and report a Dice coefficient of 0.8727 against ground truth segmentations in our dataset and statistically significant improvements in both left ventricular segmentation accuracy and image intensity estimates.

  PublisherOA PDFDOI

• Propulsion system design and energy optimization for autonomous underground freight transportation systems

  Tunnelling and Underground Space Technology · 2019-04-04 · 15 citations

  articleSenior author
  PublisherDOI

• Design and Operation of Autonomous Underground Freight Transportation Systems

  Journal of Pipeline Systems Engineering and Practice · 2019-08-23 · 12 citations

  articleSenior author

  Despite the increases in freight transportation demand, options for increasing capacity of the overground freight transportation infrastructure system are limited. This paper investigated the design and operation of an underground freight transportation (UFT) system that uses space below highways. Underground freight transportation is a class of automated transportation system in which individual vehicles carry freight through tunnels and pipelines between intermodal terminals. This paper presents attributes for schematic designs and operations for two UFT scenarios: a long-haul system which transports standard shipping containers between the Port of Houston and a terminal near Dallas, and a short-haul system which carries pallet-size freight between the Port of Houston and a satellite terminal near Houston. The appropriate design details were determined by the size of the freight, the types of vehicles and tunnels, and the propulsion system. In addition, operational attributes such as operating speed, headway, line capacity, and associated fleet sizes were addressed. Design sketches and operational equations presented in this paper are generally independent of the freight sizes and route lengths and this case study can be used as guidance for the design and operation of other freight tunnels and pipelines.

  PublisherDOI

• Life-cycle cost analysis of a short-haul underground freight transportation system for the DFW Airport

  Built Environment Project and Asset Management · 2019-05-31 · 13 citations

  articleSenior author

  Purpose The purpose of this paper is to conduct life-cycle cost analysis of a short-haul underground freight transportation (UFT) system for the Dallas Fort Worth international airport. Design/methodology/approach The research approach includes: identifying the cost components of the proposed airport UFT system; estimating life-cycle cost (LCC) of system components using various methods; determining life-cycle cash flows; evaluating the reliability of the results using sensitivity analysis; and assessing the validity of the results using analogues cases. Findings Although the capital cost of constructing an airport UFT system seems to be the largest cost of such innovative projects, annual costs for running the system are more significant, taking a life-cycle perspective. System administrative cost, tunnel operation and maintenance, and tunnel construction cost are the principle cost components of the UFT system representing approximately 46, 24 and 19 percent of the total LCC, respectively. The shipping cost is estimated to be $4.14 per ton-mile. Although this cost is more than the cost of transporting cargos by trucks, the implementation of UFT systems could be financially justified considering their numerous benefits. Originality/value This paper, for the first time, helps capital planners understand the LCC of an airport UFT system with no or limited past experience, and to consider such innovative solutions to address airport congestion issues.

  PublisherDOI

• Modeling Framework to Identify an Affected Area for Developing Traffic Management Strategies

  Journal of Transportation Engineering Part A Systems · 2018-07-31 · 4 citations

  article

  When a traffic incident occurs, congestion starts to disseminate around the incident location. Considering a suitable area to assess the impact of incidents and develop traffic network prediction models for evaluating traffic management schemes remains a challenging question. This study aims at developing a modeling framework to identify an affected area around the incident. For this purpose, linear regression models are presented to predict the maximum distance from a closed link to a link with a specified expected increase in travel time. Nine different models are presented to investigate the effects of the network topology and demand on the size of the affected area around the disruption. The models demonstrate that traffic volume on the closed link, a link’s area type, and the travel time on the first and second alternate paths with lowest travel times predict the radius of the affected area. This study will help traffic network managers reduce the complexity of their models by allowing them to use a subnetwork instead of the entire network.

  PublisherDOI

• Evaluation of Underground Freight Transportation in Texas

  Transportation Research Board 97th Annual MeetingTransportation Research Board · 2018-01-01

  article
  Publisher

• Intermodal Terminal Design for Autonomous Freight Transportation Systems

  American Journal of Transportation and Logistics · 2018-01-01

  article

  This paper investigates the design and operation of an intermodal terminal for Underground Freight Transportation (UFT) system. UFT is considered as a new mode of freight transportation that uses pipelines for transporting freight between two intermodal terminals. The load size and route length of the UFT system can be variable depending on the purpose and specifications of the project. In this paper two sizes of loads and two route lengths are considered to show that the terminal design and operation are independent from route length and load size. Each UFT route starts at the Port of Houston where shipping container or pallet size loads will be delivered to the UFT system. The short-haul route ends at a proposed intermodal satellite terminal outside Houston and the long-haul route terminates in a proposed intermodal inland port in Dallas. This paper develops inclusive equations to estimate the operational attributes of the UFT intermodal terminal. These attributes include operational headway, system flow, fleet size, and number of handlers/forklifts required in operation of the terminal. Based on the terminal operation requirements, a typical design for a UFT intermodal terminal is presented. In addition, the loading/unloading process and the freight circulation scheme are discussed. Although, this schematic terminal design and operation are for shipping container loads, but the same concept can be applied to smaller scale loads such as pallets.

  PublisherDOI

• Design and Operation of Autonomous Freight Transportation Systems

  SSRN Electronic Journal · 2018-01-01

  articleOpen accessSenior author
  PublisherDOI

Recent grants

• NIH Grant R21HL109968

  NIH · $447k · 2014

• Computational Assessment of Galectin-3 Significance in Heart Failure Remodeling

  NIH · $3.0M · 2016–2023

Frequent coauthors

• Usha Sinha

  San Diego State University

  16 shared

• Stephen Mattingly

  The University of Texas at Arlington

  11 shared

• James C. Williams

  Indiana University – Purdue University Indianapolis

  10 shared

• Sirwan Shahooei

  The University of Texas at Arlington

  9 shared

• Michael I. Miller

  Discovery Institute

  8 shared

• Robert G. Weiss

  Johns Hopkins University

  6 shared

• Robert K Herman

  6 shared

• Shekhar Govind

  6 shared

Awards & honors

• Johns Hopkins Discovery Awards (2020)