-
Investigating Why Clinicians Deviate from Standards of Care: Liberating Patients from Mechanical Ventilation in the ICU
Authors:
Nur Yildirim,
Susanna Zlotnikov,
Aradhana Venkat,
Gursimran Chawla,
Jennifer Kim,
Leigh A. Bukowski,
Jeremy M. Kahn,
James McCann,
John Zimmerman
Abstract:
Clinical practice guidelines, care pathways, and protocols are designed to support evidence-based practices for clinicians; however, their adoption remains a challenge. We set out to investigate why clinicians deviate from the ``Wake Up and Breathe'' protocol, an evidence-based guideline for liberating patients from mechanical ventilation in the intensive care unit (ICU). We conducted over 40 hour…
▽ More
Clinical practice guidelines, care pathways, and protocols are designed to support evidence-based practices for clinicians; however, their adoption remains a challenge. We set out to investigate why clinicians deviate from the ``Wake Up and Breathe'' protocol, an evidence-based guideline for liberating patients from mechanical ventilation in the intensive care unit (ICU). We conducted over 40 hours of direct observations of live clinical workflows, 17 interviews with frontline care providers, and 4 co-design workshops at three different medical intensive care units. Our findings indicate that unlike prior literature suggests, disagreement with the protocol is not a substantial barrier to adoption. Instead, the uncertainty surrounding the application of the protocol for individual patients leads clinicians to deprioritize adoption in favor of tasks where they have high certainty. Reflecting on these insights, we identify opportunities for technical systems to help clinicians in effectively executing the protocol and discuss future directions for HCI research to support the integration of protocols into clinical practice in complex, team-based healthcare settings.
△ Less
Submitted 20 February, 2024;
originally announced February 2024.
-
Sketching AI Concepts with Capabilities and Examples: AI Innovation in the Intensive Care Unit
Authors:
Nur Yildirim,
Susanna Zlotnikov,
Deniz Sayar,
Jeremy M. Kahn,
Leigh A. Bukowski,
Sher Shah Amin,
Kathryn A. Riman,
Billie S. Davis,
John S. Minturn,
Andrew J. King,
Dan Ricketts,
Lu Tang,
Venkatesh Sivaraman,
Adam Perer,
Sarah M. Preum,
James McCann,
John Zimmerman
Abstract:
Advances in artificial intelligence (AI) have enabled unprecedented capabilities, yet innovation teams struggle when envisioning AI concepts. Data science teams think of innovations users do not want, while domain experts think of innovations that cannot be built. A lack of effective ideation seems to be a breakdown point. How might multidisciplinary teams identify buildable and desirable use case…
▽ More
Advances in artificial intelligence (AI) have enabled unprecedented capabilities, yet innovation teams struggle when envisioning AI concepts. Data science teams think of innovations users do not want, while domain experts think of innovations that cannot be built. A lack of effective ideation seems to be a breakdown point. How might multidisciplinary teams identify buildable and desirable use cases? This paper presents a first hand account of ideating AI concepts to improve critical care medicine. As a team of data scientists, clinicians, and HCI researchers, we conducted a series of design workshops to explore more effective approaches to AI concept ideation and problem formulation. We detail our process, the challenges we encountered, and practices and artifacts that proved effective. We discuss the research implications for improved collaboration and stakeholder engagement, and discuss the role HCI might play in reducing the high failure rate experienced in AI innovation.
△ Less
Submitted 20 February, 2024;
originally announced February 2024.
-
Ignore, Trust, or Negotiate: Understanding Clinician Acceptance of AI-Based Treatment Recommendations in Health Care
Authors:
Venkatesh Sivaraman,
Leigh A. Bukowski,
Joel Levin,
Jeremy M. Kahn,
Adam Perer
Abstract:
Artificial intelligence (AI) in healthcare has the potential to improve patient outcomes, but clinician acceptance remains a critical barrier. We developed a novel decision support interface that provides interpretable treatment recommendations for sepsis, a life-threatening condition in which decisional uncertainty is common, treatment practices vary widely, and poor outcomes can occur even with…
▽ More
Artificial intelligence (AI) in healthcare has the potential to improve patient outcomes, but clinician acceptance remains a critical barrier. We developed a novel decision support interface that provides interpretable treatment recommendations for sepsis, a life-threatening condition in which decisional uncertainty is common, treatment practices vary widely, and poor outcomes can occur even with optimal decisions. This system formed the basis of a mixed-methods study in which 24 intensive care clinicians made AI-assisted decisions on real patient cases. We found that explanations generally increased confidence in the AI, but concordance with specific recommendations varied beyond the binary acceptance or rejection described in prior work. Although clinicians sometimes ignored or trusted the AI, they also often prioritized aspects of the recommendations to follow, reject, or delay in a process we term "negotiation." These results reveal novel barriers to adoption of treatment-focused AI tools and suggest ways to better support differing clinician perspectives.
△ Less
Submitted 31 January, 2023;
originally announced February 2023.
-
High-efficiency and broadband electro-optic frequency combs enabled by coupled micro-resonators
Authors:
Yaowen Hu,
Mengjie Yu,
Brandon Buscaino,
Neil Sinclair,
Di Zhu,
Rebecca Cheng,
Amirhassan Shams-Ansari,
Linbo Shao,
Mian Zhang,
Joseph M. Kahn,
Marko Loncar
Abstract:
Developments in integrated photonics have led to stable, compact, and broadband comb generators that support a wide range of applications. Current on-chip comb generators, however, are still limited by low optical pump-to-comb conversion efficiencies. Here, we demonstrate an integrated electro-optic frequency comb with a conversion efficiency of 30% and an optical bandwidth of 132 nm, featuring a…
▽ More
Developments in integrated photonics have led to stable, compact, and broadband comb generators that support a wide range of applications. Current on-chip comb generators, however, are still limited by low optical pump-to-comb conversion efficiencies. Here, we demonstrate an integrated electro-optic frequency comb with a conversion efficiency of 30% and an optical bandwidth of 132 nm, featuring a 100-times higher conversion efficiency and 2.2-times broader optical bandwidth compared with previous state-of-the-art integrated electro-optic combs. We further show that, enabled by the high efficiency, the device acts as an on-chip femtosecond pulse source (336 fs pulse duration), which is important for applications in nonlinear optics, sensing, and computing. As an example, in the ultra-fast and high-power regime, we demonstrate the observation of a combined EO-χ^(3) nonlinear frequency comb. Our device paves the way for practical optical frequency comb generators enabling energy-efficient computing, communication, and metrology, and provides a platform to investigate new regimes of optical physics that simultaneously involve multiple nonlinearities.
△ Less
Submitted 16 December, 2021; v1 submitted 29 November, 2021;
originally announced November 2021.
-
Improved spot formation for flexible multi-mode fiber endoscope using partial reflector
Authors:
Ruo Yu Gu,
Elaine Chou,
Cory Rewcastle,
Ofer Levi,
Joseph M. Kahn
Abstract:
Multi-mode fiber (MMF) endoscopes are a new type of endoscope that use only a single optical fiber to transmit images, achieving much higher spatial resolution for a given diameter than commercial fiber bundle endoscopes. However, a drawback of MMF endoscopes is that imaging quality degrades substantially as the MMF is perturbed. We propose a method for improving spot formation at the distal end o…
▽ More
Multi-mode fiber (MMF) endoscopes are a new type of endoscope that use only a single optical fiber to transmit images, achieving much higher spatial resolution for a given diameter than commercial fiber bundle endoscopes. However, a drawback of MMF endoscopes is that imaging quality degrades substantially as the MMF is perturbed. We propose a method for improving spot formation at the distal end of a perturbed MMF, and thus improving imaging quality, by attaching a partial reflector to the distal end of the MMF. We experimentally find that the perturbation of the light reflected from the partial reflector is highly correlated with the perturbation of the light transmitted through the MMF. We demonstrate a simple method exploiting this correlation that enables formation of spots at the distal fiber end with quality 10-15% higher than if the perturbation of the MMF is ignored. In the future, more advanced algorithms exploiting the correlation may enable further improvements in spot formation and imaging through perturbed fibers.
△ Less
Submitted 19 May, 2018;
originally announced May 2018.
-
Importance of Amplifier Physics in Maximizing the Capacity of Submarine Links
Authors:
Jose Krause Perin,
Joseph M. Kahn,
John D. Downie,
Jason Hurley,
Kevin Bennett
Abstract:
The throughput of submarine transport cables is approaching fundamental limits imposed by amplifier noise and Kerr nonlinearity. Energy constraints in ultra-long submarine links exacerbate this problem, as the throughput per fiber is further limited by the electrical power available to the undersea optical amplifiers. Recent works have studied how employing more spatial dimensions can mitigate the…
▽ More
The throughput of submarine transport cables is approaching fundamental limits imposed by amplifier noise and Kerr nonlinearity. Energy constraints in ultra-long submarine links exacerbate this problem, as the throughput per fiber is further limited by the electrical power available to the undersea optical amplifiers. Recent works have studied how employing more spatial dimensions can mitigate these limitations. In this paper, we address the fundamental question of how to optimally use each spatial dimension. Specifically, we discuss how to optimize the channel power allocation in order to maximize the information-theoretic capacity under an electrical power constraint. Our formulation accounts for amplifier physics, Kerr nonlinearity, and power feed constraints. Whereas recent works assume the optical amplifiers operate in deep saturation, where power-conversion efficiency (PCE) is high, we show that given a power constraint, operating in a less saturated regime, where PCE is lower, supports a wider bandwidth and a larger number of spatial dimensions, thereby maximizing capacity. This design strategy increases the capacity of submarine links by about 70% compared to the theoretical capacity of a recently proposed high-capacity system.
△ Less
Submitted 18 February, 2019; v1 submitted 14 March, 2018;
originally announced March 2018.
-
Low-complexity implementation of convex optimization-based phase retrieval
Authors:
Sercan O. Arik,
Joseph M. Kahn
Abstract:
Phase retrieval has important applications in optical imaging, communications and sensing. Lifting the dimensionality of the problem allows phase retrieval to be approximated as a convex optimization problem in a higher-dimensional space. Convex optimization-based phase retrieval has been shown to yield high accuracy, yet its low-complexity implementation has not been explored. In this paper, we s…
▽ More
Phase retrieval has important applications in optical imaging, communications and sensing. Lifting the dimensionality of the problem allows phase retrieval to be approximated as a convex optimization problem in a higher-dimensional space. Convex optimization-based phase retrieval has been shown to yield high accuracy, yet its low-complexity implementation has not been explored. In this paper, we study three fundamental approaches for its low-complexity implementation: the projected gradient method, the Nesterov accelerated gradient method, and the alternating direction method of multipliers (ADMM) method. We derive the corresponding estimation algorithms and evaluate their complexities. We compare their performance in the application area of direct-detection mode-division multiplexing. We demonstrate that they yield negligible estimation penalties (less than 0.2 dB for transmitter processing and less than 0.6 dB for receiver equalization) while yielding low computational cost, as their implementation complexities all scale quadratically in the number of unknown parameters. Among the three methods, ADMM achieves convergence after the smallest number of iterations.
△ Less
Submitted 19 March, 2018; v1 submitted 18 July, 2017;
originally announced July 2017.
-
Space-Time Reference with an Optical Link
Authors:
Paul Berceau,
Michael Taylor,
Joseph M. Kahn,
Leo Hollberg
Abstract:
We describe a method for realizing a high-performance Space-Time Reference (STR) using a stable atomic clock in a precisely defined orbit and synchronizing the orbiting clock to high-accuracy atomic clocks on the ground. The synchronization would be accomplished using a two-way lasercom link between ground and space. The basic concept is to take advantage of the highest-performance cold-atom atomi…
▽ More
We describe a method for realizing a high-performance Space-Time Reference (STR) using a stable atomic clock in a precisely defined orbit and synchronizing the orbiting clock to high-accuracy atomic clocks on the ground. The synchronization would be accomplished using a two-way lasercom link between ground and space. The basic concept is to take advantage of the highest-performance cold-atom atomic clocks at national standards laboratories on the ground and to transfer that performance to an orbiting clock that has good stability and that serves as a "frequency-flywheel" over time-scales of a few hours. The two-way lasercom link would also provide precise range information and thus precise orbit determination (POD). With a well-defined orbit and a synchronized clock, the satellite cold serve as a high-accuracy Space-Time Reference, providing precise time worldwide, a valuable reference frame for geodesy, and independent high-accuracy measurements of GNSS clocks. With reasonable assumptions, a practical system would be able to deliver picosecond timing worldwide and millimeter orbit determination.
△ Less
Submitted 7 December, 2015;
originally announced December 2015.
-
Wavelength-Selective Switches for Mode-Division Multiplexing: Scaling and Performance Analysis
Authors:
Keang-Po Ho,
Joseph M. Kahn,
Jeffrey P. Wilde
Abstract:
Wavelength-selective switches for mode-division-multiplexing systems are designed by scaling switches from single-mode systems. All modes at a given wavelength are switched as a unit, which is necessary in systems with substantial mode coupling, and minimizes the number of ports required to accommodate a given traffic volume. When a pure mode is present at the input, modal transmission and couplin…
▽ More
Wavelength-selective switches for mode-division-multiplexing systems are designed by scaling switches from single-mode systems. All modes at a given wavelength are switched as a unit, which is necessary in systems with substantial mode coupling, and minimizes the number of ports required to accommodate a given traffic volume. When a pure mode is present at the input, modal transmission and coupling coefficients are mode-dependent and may be computed using a simple mode-clip** model. When multiple modes are present, interference between modes alters the transmission and coupling coefficients, shifting the passband center frequency and changing its bandwidth. Mode-coupling matrices are used to compute mixed modes having the narrowest or widest bandwidths, or having the largest center-frequency offsets. In a specific design for graded-index fiber, five mode groups and 50-GHz channel spacing, the one-sided bandwidth may change up to $\pm$3.6 GHz. In a system with many cascaded switches and strong mode coupling, the end-to-end response per switch may be characterized by a mode-averaged transmission coefficient.
△ Less
Submitted 4 September, 2014; v1 submitted 28 January, 2014;
originally announced January 2014.
-
Frequency Diversity in Mode-Division Multiplexing Systems
Authors:
Keang-Po Ho,
Joseph M. Kahn
Abstract:
In the regime of strong mode coupling, the modal gains and losses and the modal group delays of a multimode fiber are known to have well-defined statistical properties. In mode-division multiplexing, mode-dependent gains and losses are known to cause fluctuations in the channel capacity, so that the capacity at finite outage probability can be substantially lower than the average capacity. Mode-de…
▽ More
In the regime of strong mode coupling, the modal gains and losses and the modal group delays of a multimode fiber are known to have well-defined statistical properties. In mode-division multiplexing, mode-dependent gains and losses are known to cause fluctuations in the channel capacity, so that the capacity at finite outage probability can be substantially lower than the average capacity. Mode-dependent gains and losses, when frequency-dependent, have a coherence bandwidth that is inversely proportional to the modal group delay spread. When mode-division-multiplexed signals occupy a bandwidth far larger than the coherence bandwidth, the mode-dependent gains and losses are averaged over frequency, causing the outage capacity to approach the average capacity. The difference between the average and outage capacities is found to be inversely proportional to the square-root of a diversity order that is given approximately by the ratio of the signal bandwidth to the coherence bandwidth.
△ Less
Submitted 23 August, 2011;
originally announced August 2011.
-
Mode-Dependent Loss and Gain: Statistics and Effect on Mode-Division Multiplexing
Authors:
Keang-Po Ho,
Joseph M. Kahn
Abstract:
In multimode fiber transmission systems, mode-dependent loss and gain (collectively referred to as MDL) pose fundamental performance limitations. In the regime of strong mode coupling, the statistics of MDL (expressed in decibels or log power gain units) can be described by the eigenvalue distribution of zero-trace Gaussian unitary ensemble in the small-MDL region that is expected to be of interes…
▽ More
In multimode fiber transmission systems, mode-dependent loss and gain (collectively referred to as MDL) pose fundamental performance limitations. In the regime of strong mode coupling, the statistics of MDL (expressed in decibels or log power gain units) can be described by the eigenvalue distribution of zero-trace Gaussian unitary ensemble in the small-MDL region that is expected to be of interest for practical long-haul transmission. Information-theoretic channel capacities of mode-division-multiplexed systems in the presence of MDL are studied, including average and outage capacities, with and without channel state information.
△ Less
Submitted 18 May, 2011;
originally announced May 2011.
-
Statistics of Group Delays in Multimode Fiber with Strong Mode Coupling
Authors:
Keang-Po Ho,
Joseph M. Kahn
Abstract:
The modal group delays (GDs) are a key property governing the dispersion of signals propagating in a multimode fiber (MMF). A MMF is in the strong-coupling regime when the total length of the MMF is much greater than the correlation length over which local principal modes can be considered constant. In this regime, the GDs can be described as the eigenvalues of zero-trace Gaussian unitary ensemble…
▽ More
The modal group delays (GDs) are a key property governing the dispersion of signals propagating in a multimode fiber (MMF). A MMF is in the strong-coupling regime when the total length of the MMF is much greater than the correlation length over which local principal modes can be considered constant. In this regime, the GDs can be described as the eigenvalues of zero-trace Gaussian unitary ensemble, and the probability density function (p.d.f.) of the GDs is the eigenvalue distribution of the ensemble. For fibers with two to seven modes, the marginal p.d.f. of the GDs is derived analytically. For fibers with a large number of modes, this p.d.f. is shown to approach a semicircle distribution. In the strong-coupling regime, the delay spread is proportional to the square root of the number of independent sections, or the square root of the overall fiber length.
This revision also made clarification to the original paper, the group delay statistics is also extended to high number of modes.
△ Less
Submitted 15 April, 2012; v1 submitted 22 April, 2011;
originally announced April 2011.
-
Hiding Single Photons With Spread Spectrum Technology
Authors:
Chinmay Belthangady,
Chih-Sung Chuu,
Ite A. Yu,
G. Y. Yin,
J. M. Kahn,
S. E. Harris
Abstract:
We describe a proof-of-principal experiment demonstrating the use of spread spectrum technology at the single photon level. We show how single photons with a prescribed temporal shape, in the presence of interfering noise, may be hidden and recovered.
We describe a proof-of-principal experiment demonstrating the use of spread spectrum technology at the single photon level. We show how single photons with a prescribed temporal shape, in the presence of interfering noise, may be hidden and recovered.
△ Less
Submitted 1 April, 2010;
originally announced April 2010.
-
Electronic Compensation Technique to Mitigate Nonlinear Phase Noise
Authors:
Keang-Po Ho,
Joseph M. Kahn
Abstract:
Nonlinear phase noise, often called the Gordon-Mollenauer effect, can be compensated electronically by subtracting from the received phase a correction proportional to the received intensity. The optimal scaling factor is derived analytically and found to be approximately equal to half of the ratio of mean nonlinear phase noise and the mean received intensity. Using optimal compensation, the sta…
▽ More
Nonlinear phase noise, often called the Gordon-Mollenauer effect, can be compensated electronically by subtracting from the received phase a correction proportional to the received intensity. The optimal scaling factor is derived analytically and found to be approximately equal to half of the ratio of mean nonlinear phase noise and the mean received intensity. Using optimal compensation, the standard deviation of residual phase noise is halved, doubling the transmission distance in systems limited by nonlinear phase noise.
△ Less
Submitted 29 May, 2003; v1 submitted 21 November, 2002;
originally announced November 2002.
