University of Central Florida

The main objectives of this study were to 1) review the literature on the applications of soft computing concepts to the field of human factors and ergonomics (HFE) between 2013 and 2022 and 2) highlight future developments and trends. Multiple soft computing methods and techniques have been investigated for their ability to address various applications in HFE effectively. These techniques include fuzzy logic, artificial neural networks, genetic algorithms, and their combinations. Applications of these methods in HFE have been highlighted in one hundred and four articles selected from 406 papers. The results of this study help address the challenges of complexity, vagueness, and imprecision in human factors and ergonomics research through the application of soft computing methodologies.

We consider a network of N sensors, tasked with solving binary distributed detection, a fusion center (FC), and a feedback channel from the FC to sensors. Each sensor is capable of harvesting energy and is equipped with a finite-size battery to store randomly arrived energy. Sensors process their observations and transmit their symbols to the FC over orthogonal and Markovian time correlated fading channels. The FC fuses the received symbols and makes a global binary decision. We aim at developing adaptive channel-dependent transmit power control policies such that J-divergence based detection metric is maximized at the FC, subject to total transmit power constraint. Modeling quantized fading channel, energy arrival, and battery dynamics as time-homogeneous finite-state Markov chains, and the network lifetime as a geometric random variable, we formulate our power control optimization problem as a discounted infinite-horizon constrained Markov decision process (MDP) problem, where sensors’ transmit powers are functions of the battery states, quantized channel gains, and the arrived energies. We utilize stochastic dynamic programming and Lagrangian approach to find the optimal and sub-optimal power control policies. We demonstrate that our sub-optimal policy provides a close-to-optimal performance with a reduced computational complexity and without imposing signaling overhead on sensors.

Mobile user profiling refers to the efforts of extracting users’ characteristics from mobile activities. In order to capture the dynamic varying of user characteristics for generating effective user profiling, we propose an imitation-based mobile user profiling framework. Considering the objective of teaching an autonomous agent to imitate user mobility based on the user’s profile, the user profile is the most accurate when the agent can perfectly mimic the user behavior patterns. The profiling framework is formulated into a reinforcement learning task, where an agent is a next-visit planner, an action is a POI that a user will visit next, and the state of the environment is a fused representation of a user and spatial entities. An event in which a user visits a POI will construct a new state, which helps the agent predict users’ mobility more accurately. In the framework, we introduce a spatial Knowledge Graph (KG) to characterize the semantics of user visits over connected spatial entities. Additionally, we develop a mutual-updating strategy to quantify the state that evolves over time. Along these lines, we develop a reinforcement imitative graph learning framework for mobile user profiling. Finally, we conduct extensive experiments to demonstrate the superiority of our approach.

Background Omadacycline (OMC), the first aminomethylcycline antibiotic, is a semisynthetic tetracycline derivative approved for community-acquired bacterial pneumonia (CABP) and acute bacterial soft skin infection (ABSSI), with a broad range of activity against Gram-positive and Gram-negative aerobes, anaerobes and atypical bacteria, including drug-resistant strains. It is currently under development for the treatment of cystitis and acute pyelonephritis (AP). Methods This systematic review of literature followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. Abstracts were searched using the term: “omadacycline”. The electronic research literature databases included the Cochrane Library, MedLine, and clinicaltrials.gov from February to April 2023. Selection of studies by PRISMA flowchart is shown in figure 1. Characteristics of endpoints for each pathology are mentioned in table 1.Table 1.Endpoints for each disease. CABP (Community-acquired Pneumonia), ECR (Early clinical response), Investigator-assessed clinical response (IACR), Post treatment evaluation (PTE), ABSSI (Acute bacterial soft skin infection), EOT, (End of treatment), AP (Acute pyelonephritis). Results OMC was found to be noninferior to existing antibiotics in all reviewed studies. The most common side effects of OMC were nausea, vomiting and diarrhea, which were infrequent, mild and easily treatable. The only limitation of these studies was proving the efficacy of OMC by non-inferiority. Bias was assessed using the Cochrane collaboration's tool risk assessment of the clinical trials (Figure 2). Characteristics and outcomes of each study are mentioned in table 2.Table 2.Characteristics, outcomes and limitations of selected studies. CABP (Community-acquired Pneumonia), OMC (Omadacycline), IV (Intravenous), ECR (Early clinical response, IACR (Investigator-assessed clinical response), PTE (Post-treatment evaluation), AE (Adverse effect), ABSSI (Acute bacterial soft skin infection), EOT (End of treatment), AP (Acute pyelonephritis), and LEV (Levofloxacin). References Conclusion Most studies showed that OMC was well tolerated, with decreased frequency of dosing, availability of both oral and intravenous options, and with equal if not better efficacy than the existing drugs. OMC has activity against common tetracycline resistance mechanisms such as efflux pumps and ribosomal protection proteins. It is excreted unchanged, having low potential for drug-drug interaction and requires no dose adjustment for age, sex, or hepatic and renal impairment. Further large scale studies are encouraged to prove OMC as an emerging antibiotic of choice for several bacterial infections. Overall, OMC appears to be a promising drug with a good safety profile. With the help of this systematic review, we aim to encourage the use of OMC for cystitis and pyelonephritis, along with continued use for CABP and ABSSI. Disclosures All Authors: No reported disclosures

Background Lenacapavir (LEN) is a highly potent, long-acting, first-in-class inhibitor of HIV-1 capsid protein for the treatment of HIV-1 infection in adults with multidrug resistance (MDR) in combination with other antiretrovirals. Methods CAPELLA is an ongoing, phase 2/3 study in heavily treatment-experienced people with HIV-1 with multidrug-resistance. Participants received oral LEN for loading followed by subcutaneous LEN Q6M in addition to an investigator-selected optimized background regimen (OBR). After reporting the efficacy and safety data through 52 weeks (W), the protocol was amended to allow longer follow up; we report W104 results. Results Of 72 participants enrolled (36 in each cohort), 25% were female, 38% were Black, median age was 52 years, 64% had CD4 <200 cells/µL, and 46% had HIV-1 resistant to all 4 major classes (NRTI, NNRTI, PI, INSTI). One participant (of 72) decided not to continue in the post-W52 extension. At W104, 62% (44/71) had HIV-1 RNA <50 c/mL via FDA Snapshot algorithm, and 14% (10/71) had HIV-1 RNA >50 c/mL; 24% discontinued for reasons other than lack of efficacy (13/71) or had missing data but were on study drug (4/71). When analyzed as missing=excluded, 81% (44/54) had HIV-1 RNA <50 c/mL. CD4 count increased by a median 97 cells/µL (Q1 to Q3: 18 to 224) and the proportion of participants with CD4 count ≥200 cells/µL increased from 36% (26/72) at baseline to 71% (39/55) at W104. Fourteen participants had emergent LEN resistance (9 previously reported), 6 of whom subsequently suppressed while continuing LEN. The median (Q1–Q3) duration of follow-up on LEN was 125 (111–140) weeks. One participant discontinued due to injection site nodule at W52 (Grade 1, previously reported); no participant discontinued due to an AE after W52. Most common AEs (excluding injection site reactions [ISRs] and COVID-19) were diarrhea and nausea (19% each, +5% each since W52, respectively). LEN-related ISRs were mostly mild-to-moderate. Conclusion In people with MDR HIV-1 and limited treatment options, LEN in combination with an OBR was well tolerated and maintained virologic suppression at W104. These data support the use of LEN for treatment of people with MDR HIV-1. Disclosures Onyema Ogbuagu, MD, Gilead Sciences, Inc: Advisor/Consultant|Gilead Sciences, Inc: Honoraria|Janssen: Advisor/Consultant|Viiv: Advisor/Consultant Edwin DeJesus, MD, Gilead Sciences, Inc: Advisor/Consultant|Theratechnology: Advisor/Consultant|Theratechnology: Honoraria Mezgebe Berhe, MD, MPH, Gilead Sciences, Inc: Clinical Trial Investigator Gary J Richmond, MD, Gilead Sciences, Inc: Grant/Research Support|Glaxo Smith Kline: Grant/Research Support|Insmed: Grant/Research Support|TaiMed: Grant/Research Support Peter J Ruane, MD, Gilead Sciences, Inc: Advisor/Consultant|Viiv: Advisor/Consultant|Viiv: Honoraria Gary I. Sinclair, MD, Abbvie: Grant/Research Support|Gilead: Advisor/Consultant|Gilead: Grant/Research Support|Janssen: Advisor/Consultant|Janssen: Grant/Research Support|Janssen: Honoraria|Merck: Advisor/Consultant|Merck: Grant/Research Support|Merck: Honoraria|Thera: Advisor/Consultant|Thera: Grant/Research Support|Thera: Honoraria|ViiV: Advisor/Consultant|ViiV: Grant/Research Support|ViiV: Honoraria Moti N Ramgopal, MD, Abbvie: Honoraria|Gilead Sciences, Inc: Advisor/Consultant|Gilead Sciences, Inc: Honoraria|Janssen: Advisor/Consultant|Janssen: Honoraria|Merck: Advisor/Consultant|Viiv: Advisor/Consultant|Viiv: Honoraria William Sanchez, MD, Gilead Sciences, Inc: Grant/Research Support|GSK: Grant/Research Support|Merck: Advisor/Consultant|Merck: Grant/Research Support Gordon E Crofoot, MD, Gilead Sciences, Inc: Clinical Trial Investigator Nicolas A Margot, MA, Gilead Sciences, Inc: Employee|Gilead Sciences, Inc: Stocks/Bonds Hui Wang, PhD, Gilead Sciences, Inc: Employee|Gilead Sciences, Inc: Stocks/Bonds Hadas Dvory-Sobol, PhD, Gilead Sciences, Inc: Employee|Gilead Sciences, Inc: Stocks/Bonds Martin Rhee, MD, Gilead Sciences, Inc: Employee|Gilead Sciences, Inc: Stocks/Bonds Jared Baeten, MD, PhD, Gilead Sciences, Inc: Employee|Gilead Sciences, Inc: Stocks/Bonds Sorana Segal-Maurer, MD, Gilead Sciences, Inc: Advisor/Consultant|Gilead Sciences, Inc: Honoraria|Janssen: Advisor/Consultant|Janssen: Honoraria|Theratechnologies: Advisor/Consultant|Theratechnologies: Honoraria|Viiv: Advisor/Consultant|Viiv: Honoraria

Background Outpatient Parenteral Antibiotic Therapy (OPAT) involves administering antibiotics outside hospitals via non-oral routes. Success depends on well-reasoned selection of patients for this treatment modality and proper antibiotic stewardship to avoid serious complications. Methods We created a de-identified dataset (n=295) from a retrospective chart review of veterans receiving OPAT at the Orlando VA Medical Center (ORL-VAMC) in 2020. We included patients aged 18 or older who underwent OPAT therapy at ORL-VAMC with at least one antibiotic administration between January 1 and December 31, 2020. Inclusion also required compliance with follow-up and laboratory monitoring for at least 90 days after starting outpatient antibiotics. Of 364 patients receiving OPAT during 2020, 295 met the full criteria, with loss to follow-up and absent labs as common reasons for exclusion. We examined patient demographics, medical history, diagnosis, and elements of treatment for associations with unplanned hospital visits within 90-days of OPAT initiation using Binary Logistic Regression modeling. Results The average age within the dataset was 67.37 (SD=±11.38) years. Common treatment indications included Osteomyelitis (80/295, 27.12%) and Bacteremia (64/295, 21.69%). 40.96% (119/295) of patients had an unplanned hospital visit within 90 days, primarily due to unresolved symptoms. Our regression model identified Pyelonephritis/UTI (OR 2.26; 95% CI 1.02-5.41), Intra-Abdominal Infection (OR 3.81, 95% CI 1.15-17.45), Cephalosporin Antibiotics (OR 2.57, 1.39-4.91), Liver Disease (OR 2.26, 95% CI 1.02-5.41), and psychiatric illness (OR 1.74, 95% CI 1.06-2.87) as being significantly associated with unplanned hospital visits within 90 days. Conclusion A majority (177/295, 60%) of patients did not need further care in the hospital setting after starting OPAT. Our study supports OPAT as an effective option for delivering IV antibiotic therapy when indicated. This study's external validity is limited by the overrepresentation of males in the sample. Nevertheless, these findings can help physicians better anticipate successful outcomes and select suitable patients for OPAT. Further research aimed at understanding the associations identified may offer improved clinical utility. Disclosures All Authors: No reported disclosures

Background In TANGO, switching to dolutegravir/lamivudine (DTG/3TC) was noninferior compared to continuing a tenofovir alafenamide (TAF)-based regimen, however switching from bictegravir (BIC)/emtricitabine (F)/TAF was not evaluated. Here, we present efficacy and safety of switching to DTG/3TC compared with continuing B/F/TAF in virologically suppressed adults through 24 weeks. Methods DYAD (NCT04585737) is an ongoing, open-label clinical trial that randomized adults with HIV-1 RNA< 50 copies/mL and no prior virologic failure (2:1) to switch to once-daily fixed-dose DTG/3TC or remain on B/F/TAF. The primary endpoint is the proportion with HIV-1 RNA ≥ 50 c/mL at Week (W) 48 (FDA snapshot algorithm, ITT-E population). A planned W24 interim analysis assessed noninferiority with a 6% margin. Results Overall, 222 adults (16% women; 51% aged ≥50 years; 28% Black) were randomized. At W24, 3 (2%) participants on DTG/3TC and 3 (4%) on B/F/TAF had HIV-1 RNA ≥ 50 c/mL (adjusted treatment difference -2.1%, 95% confidence interval [-9.5%, 2.5%]) meeting noninferiority criteria. Through W24, 4 on DTG/3TC and 3 on B/F/TAF met confirmed virologic withdrawal (CVW) criteria and underwent resistance testing. 2/7 had treatment-emergent resistance. One B/F/TAF CVW developed M184M/I and G140G/S at W12; at study discontinuation (DC), HIV-1 RNA< 50 c/mL and the participant remained on B/F/TAF. One DTG/3TC CVW had no resistance on study genotype but underwent repeat genotypic testing outside the study (ordered by clinic provider) and had M184V at W12; at study DC, HIV-1 RNA< 50 c/mL on DTG/3TC and the participant was subsequently transitioned to DTG + darunavir/cobicistat (DRV/c). One non-CVW DTG/3TC participant developed M184V and K65R at W12 (genotype inadvertently collected at first episode of unconfirmed viremia); at study DC, HIV-1 RNA< 50 c/mL on DTG/3TC and the participant was subsequently transitioned to DTG+DRV/c. Drug-related adverse events (AEs) and withdrawals due to AEs occurred in 31 (21%) and 6 (4%) participants with DTG/3TC and 1 (1%) and 0 participants with B/F/TAF, respectively. Conclusion DYAD demonstrated noninferior efficacy of switching to DTG/3TC vs. continuing B/F/TAF at W24. Higher AE rates in the DTG/3TC arm are likely consistent with the open-label nature of this switch study. Disclosures Charlotte-Paige M. Rolle, MD, MPH, Gilead: Advisor/Consultant|Gilead: Grant/Research Support|Gilead: Honoraria|Janssen: Advisor/Consultant|ViiV: Advisor/Consultant|ViiV: Grant/Research Support|ViiV: Honoraria Federico Hinestrosa, MD, AbbVie: Honoraria|Gilead Sciences: Advisor/Consultant|Gilead Sciences: Honoraria|MSD: Honoraria|ViiV Healthcare: Advisor/Consultant|ViiV Healthcare: Honoraria Edwin DeJesus, MD, Gilead Sciences, Inc: Advisor/Consultant|Theratechnology: Advisor/Consultant|Theratechnology: Honoraria

This chapter surveys theories of the relationship between narrative and various forms of experience by reviewing the role narrative plays in a range of academic disciplines. It begins with influential attempts to define narrative (Goldie, Nelson, Kintsch and van Dijk, Ryan, and Herman), then considers theories of narrative’s role as social discourse (Labov and Waletzky, Mink, and White), before turning to thinkers who view narrative as part of cognitive processes (MacIntyre, Mair, Hutto, Bruner, Dennett, Fludernik, and Herman). It focuses on specific theories of narrative’s relation to perception (Oatley, Schank and Abelson, Lakoff and Johnson, Holland and Kensinger, Menary, Carr, and Turner). Finally, it considers how narrative has been implicated in selfhood and how it emerges developmentally (Markowitsch and Stanilou, Dennett, Menary, Oatley, Nelson, Merrill, Fivush).

This chapter argues that narrativity permeates cognitive functions related to selfhood and the social origins of our valuations. ‘Higher order’ cognitive processes imbue our experiences with elements of weak narrativity and are in turn shaped by the weakly narrative ‘givens’ of our experiences. These include our sense of ongoing continuity (Baynes, Zahavi, Kriegel, Dretske, Prinz), the deployment of knowledge garnered from prior experiences to make sense of present experiences (McRae, Brown, and Ellmann; Ghosh and Gilboa, Bar, Arbib, McClelland), learning from present experiences, our sense(s) of selfhood (Christoff et al., Menary, Metzinger, Damasio, Deleuze, Velleman), autobiographical selfhood (Dennett, Ricouer, Freeman, Eakin, Oatley, Markowitsch and Stanilou, Rubin, Schrauf, and Greenberg, Fivush), and social cognition (Hasson, Furman et al., Hutto and McGivern, Maibom, Dewhurst, Hassabis and Maguire, Beatty). Viewing narrative from a processing perspective allows us to theorize a vital role for narrative in our experiences, identities, and social intuitions.

This conclusion reprises and explores the implications of the book’s central argument that narrative processing is part of ordinary cognition, amounting to emergent ‘weak narrativity.’ Narrative processing is ongoing meaning-directed generation and processing of emergent temporal structures consisting of related value-dynamic sequences involving value-dynamic selected agents, mediated by socio-cultural contexts. This definition applies to processes occurring in a wide range of cognitive activities; also, elements of the definition describe aspects of discourses and experience that are properly bundled together and for which ‘narrativity’ is an apt description. This view of narrative processing has implications for a wide range of approaches to literary studies, and it suggests research directions in literary studies, philosophy, and the cognitive sciences.

This chapter argues that the origins and nature of narrativity precede and exceed conventional notions of intention. That is, our intentions to construct and communicate narratives are not solely responsible for the narrative characteristics we discern in experiences. Instead, intention is shaped by intuitions of narrativity derived simultaneously from top-down (through received sociocultural values and frameworks) and bottom-up (through perceptions ready-made for narrative construction) sources. This claim supports a larger argument that weak narrativity is emergent in cognitive processing. That is, narrativity can emerge in the apprehension of events, prior to conscious intention to construct a narrative. The view complements models of narrative as ‘small,’ embedded, and iterative, and is an elaboration of conversations about narrative as a mode of thinking (De Fina and Georgakopoulou 2011, p. 15).

This chapter considers the cognitive processing that occurs during the reading of narratives. It builds on Fludernik’s concept of ‘experientiality’ to describe the abstraction processes of both ordinary experiences and narrative reading in terms of the role the felt quality of experiences plays in information processing, rather than vice versa. This leads to a taxonomy of emergent structures in reading and ordinary experiences, including emergent qualities such as relational salience and dynamic valuation. The chapter concludes by describing the bottom-up aspects of reading narrative from these experiential building blocks.

This chapter argues that weak narrativity exists in low-level cognition. Specifically, weak narrativity helps explain the phenomenology of event experiences. The chapter considers research on the phenomenology of time (Bergson, Montemayor, Wittmann), on event segmentation (Shipley, Zacks, Tversky, Bauer, Hommel, Schacter and Addis, Davachi and Dubrow), and on episodic memory (Conway, Moulin, Spreng, Mar, Kim). Narrative processing involves synthesizing key features of temporality and event segmentation: selection, concatenation, boundary marking, and dynamic valuation of agents. This synthesis activity is core to ordinary experience. It describes the ongoing tracking of value dynamics within the continuous flow of experience, and how an ongoing, Ricouerian grasping together of temporally extended value dynamics accretes into meaning-laden phenomena. The chapter concludes by situating this argument within philosophical perspectives that argue for the importance of narrative in cognition (Butler, Menary), specifically by connecting the hermeneutical tradition with recent insight into episodic memory.

This chapter briefly introduces the history of cognitive studies and several contemporary models of cognition, with an emphasis on philosophical views. It offers a primer on debates about cognition, including Global Workspace Theory (Baars, Raichle, Buckner, Andrews, Schacter, Hassabis et al.), theories of memory systems (Tulving, Conway, Piolino et al.), the default mode network, predictive processing (Friston, Metzinger and Wiese, Hohwy, Anderson), and embodied cognition (Bruineberg, Clark, Maiese, O’Donovan-Andersen, Shapiro, Thompson, Zahavi, Eilan). These models are considered in more detail in subsequent chapters. Here they are introduced and connected to the ideas about narrative articulated in the first half of the book.

This chapter argues that affect, as a form of valuation, shapes the emergent narrativity of some experiences. The chapter examines research arguing for a key role for affect in cognition (Cromwell, Panksepp, Bauer, Scherer, Ellsworth, Feldman Barrett, Oatley, Clore, Power and Dagleish, de Sousa, Damasio, Tappolet, Roberts, Prinz, Pessoa, Kahneman, Critchley and Garfinkel, Seth, Mesquita, LeDoux), as well as philosophical and literary arguments that link narrative to affect and emotion (Sternberg, Velleman, Hogan, Brown). The chapter considers several models which specify how dynamic relational networks, spatial boundaries, agent selection, and meaningful temporal structures are shaped by the affective economy of experiences via factors like arousal, valence, salience, selection, and allostatic set points. Affect is a basis for processes of dynamic, relational valuation that produce weak narrativity in experience.

We give a combinatorial interpretation of vector continued fractions obtained by applying the Jacobi–Perron algorithm to a vector of \(p\ge 1\) resolvent functions of a banded Hessenberg operator of order \(p+1\). The interpretation consists in the identification of the coefficients in the power series expansion of the resolvent functions as weight polynomials associated with Lukasiewicz lattice paths in the upper half-plane. In the scalar case \(p=1\) this reduces to the relation established by P. Flajolet and G. Viennot between Jacobi–Stieltjes continued fractions, their power series expansion, and Motzkin paths. We consider three classes of lattice paths, namely the Lukasiewicz paths in the upper half-plane, their symmetric images in the lower half-plane, and a third class of unrestricted lattice paths which are allowed to cross the x-axis. We establish a relation between the three families of paths by means of a relation between the associated generating power series. We also discuss the subcollection of Lukasiewicz paths formed by the partial p-Dyck paths, whose weight polynomials are known in the literature as genetic sums or generalized Stieltjes–Rogers polynomials, and express certain moments of bi-diagonal Hessenberg operators.

Throughout the last few decades, a significant amount of research has been conducted on the wetting and deformation of soft materials. Employing high-speed videography, we report the spatiotemporal wetting and hemiwicking velocities for different fluids with micropillar hemiwicking arrays of various pillar geometries, spacing configurations, and Young's moduli ranging from E = 61 kPa to 12 GPa. The wetting and wicking dynamics are analyzed with comparisons to that predicted by contemporary scaling laws or derived analytic solutions. The results show that our analytical models can predict the inertial wetting, deformation, and hemiwicking dynamics within typically <50% of that experimentally observed. For the wide range of sample-fluid couples studied, the upper-limit (maximum) hemiwicking and inertial wetting velocities are typically observed within the narrow ranges of 10~30 mm/s and 0.8~1.6 m/s, respectively. Aside from the ultrasoft (~61 kPa) Ecoflex sample, sample stiffness had no notable effect on the maximum speed that a fluid can wet the non-textured surface region. However, for soft materials with elastic moduli below ~150 kPa, the rapid coupling between transient swelling and elastocapillary deformations leads to recoil wetting and wicking dynamics and notable reductions in the hemiwicking velocity.

Inverse obstacle scattering is the recovery of an obstacle boundary from the scattering data produced by incident waves. This shape recovery can be done by iteratively solving a PDE-constrained optimization problem for the obstacle boundary. While it is well known that this problem is typically non-convex and ill-posed, previous investigations have shown that in many settings these issues can be alleviated by using a continuation-in-frequency method and introducing a regularization that limits the frequency content of the obstacle boundary. It has been recently observed that these techniques can fail for obstacles with pronounced cavities, even in the case of penetrable obstacles where similar optimization and regularization methods work for the equivalent problem of recovering a piecewise constant wave speed. The present work investigates the recovery of obstacle boundaries for impenetrable, sound-soft media with pronounced cavities, given multi-frequency scattering data. Numerical examples demonstrate that the problem is sensitive to the choice of iterative solver used at each frequency and the initial guess at the lowest frequency. We propose a modified continuation-in-frequency method which follows a random walk in frequency, as opposed to the standard monotonically increasing path. This method shows some increased robustness in recovering cavities, but can also fail for more extreme examples. An interesting phenomenon is observed that while the obstacle reconstructions obtained over several random trials can vary significantly near the cavity, the results are consistent for non-cavity parts of the boundary.