Wednesday, July 21, 2021

Emory team makes finals in Amazon's Alexa Prize for artificial intelligence

The Emory team is led by faculty advisor Jinho Choi (center) and graduate students James Finch (left) and Sarah Finch.

A team of six Emory computer science students made it to the final round for Amazon’s Alexa Prize Socialbot Grand Challenge, a global competition among universities to create a chatbot that advances the field of artificial intelligence. The winner of the 2021 Alexa Prize will be announced in mid-August. At stake is a $500,000 first prize. In addition, $1 million in research funds will be awarded to the winning team if it meets the “grand challenge” criteria, including the ability of its chatbot to engage the judges in conversation for at least 20 minutes. 
 
In addition to Emory, the finalists are Czech Technical University, Prague; SUNY at Buffalo, New York; Stanford University and the University of California Santa Cruz. 
 
The Emory team is headed by graduate students Sarah Finch and James Finch, along with faculty advisor Jinho Choi, assistant professor in the Department of Computer Sciences. Last year, the trio headed a team of 14 Emory students that took first place, winning $500,000 for their chatbot named Emora. They chose the name because it sounds like a feminine version of “Emory” and is similar to a Hebrew word for an eloquent sage. 

This year, they are turning up the heat with an even more advanced version of Emora and new team members, including graduate student Han He and undergraduates Sophy Huang, Daniil Huryn and Mack Hutsell. All the students are members of Choi’s Natural Language Processing Research Laboratory

“I’m extremely proud to have such a talented team of students,” Choi says. “It’s a group of strongly motivated people with the right combination of diverse skills coming together at the right time. They’re working on changing the paradigm for conversational artificial intelligence.” 

“We’re using some established technology but taking a groundbreaking approach in how we combine and execute dialogue management so a computer can make logical inferences while conversing with a human,” adds Sarah Finch. “Ultimately, we’re making Emora even more flexible in how she can interact with people.”


Related:

Monday, July 12, 2021

Using big data to zero in on a mosquito menace

"The ultimate goal is to give public health officials the power to do more effective and efficient mosquito control — even before an epidemic beings," says Gonzalo Vazquez-Prokopec, right, shown with public health technicians in Merida, Mexico.

Scientists from Emory University are leading a new strategy in humanity’s long-running war with the mosquito, using big data to target one species of this tiny disease vector. They have studied the behavior of the Aedes aegypti mosquito, pinpointing where it hangs out in homes. They are mapping cases over time of viral infections that this species can transmit when it bites a human — including dengue fever, Zika, chikungunya and yellow fever. They are zeroing in on high-risk neighborhoods for outbreaks of these diseases in sprawling urban areas of the tropics. 

“We’re working on some of the boldest and biggest changes in decades for the surveillance and control of this mosquito,” says Gonzalo Vazquez-Prokopec, associate professor in Emory's Department of Environmental Sciences
 
The National Institutes of Health’s Division of Microbiology and Infectious Diseases awarded Vazquez-Prokopec a $6.5 million grant to lead a consortium in a randomized clinical trial in Merida, Mexico, as part of the quest to reduce Aedes-borne viral infections. The trial is testing a novel intervention that previous Emory research finds promising: Indoor residual spraying of insecticide, targeted to the places in homes where this mosquito tends to rest. 
 

Related:

Wednesday, July 7, 2021

Mapping dengue hot spots pinpoints risks for Zika and chikungunya

A female Aedes aegypti mosquito takes a blood meal from a human. "The ultimate goal is to give public health officials the power to harness big data and do more effective and efficient mosquito control — even before an epidemic begins," says Emory disease ecologist Gonzalo Vazquez-Prokopec. (Photo by James Gathany / CDC).

By Carol Clark

Data from nine cities in Mexico confirms that identifying dengue fever “hot spots” can provide a predictive map for future outbreaks of Zika and chikungunya. All three of these viral diseases are spread by the Aedes aegypti mosquito. 

Lancet Planetary Health published the research, led by Gonzalo Vazquez-Prokopec, associate professor in Emory University’s Department of Environmental Sciences. The study provides a risk-stratification method to more effectively guide the control of diseases spread by Aedes aegypti

“Our results can help public health officials to do targeted, proactive interventions for emerging Aedes-borne diseases,” Vazquez-Prokopec says. “We’re providing them with statistical frameworks in the form of maps to guide their actions.” 

The study encompassed data for 2008 through 2020 from cities in southern Mexico with a high burden of dengue fever cases during that period, along with cases of the more recently emerged diseases of Zika and chikungunya. The cities included Acapulco, Merida, Veracruz, Cancun, Tapachula, Villahermosa, Campeche, Iguala and Coatzacoalcos. 

The results found a 62 percent overlap of hot spots for dengue and Zika and 53 percent overlap for cases of dengue and chikungunya. In addition, dengue hot spots between 2008 and 2016 were significantly associated with dengue hotspots detected between 2017 and 2020 in five of the nine cities. 

The work builds on a previous study of the spatial-temporal overlap of the three diseases, focused on Merida, a city of one million located in the Yucatan Peninsula. That study showed that nearly half of Merida’s dengue cases from 2008 to 2015 were clustered in 27 percent of the city. These dengue hot spots contained 75 percent of the first chikungunya cases reported during the outbreak of that disease in 2015 and 100 percent of the first Zika cases reported during the Zika outbreak of 2016. 

“In this latest paper, we’ve expanded our analysis in scope and geography and shown that the findings are consistent across these nine cities of different sizes and in different regions,” Vazquez-Prokopec says. “We’ve confirmed that dengue, Zika and chikungunya outbreaks tend to concentrate in small areas of a city, and that these hot spots are predictive of where future cases will concentrate.” 

Mosquito control efforts generally involve outdoor spraying that covers broad swaths of a city, but the Aedes aegypti mosquito has adapted to live inside houses. Work by Vazquez-Prokopec and collaborators has shown that the best way to control these mosquitos and the diseases they spread is by spraying a long-lasting pesticide indoors — on the ceilings, along the bases of walls, and in other areas of homes where the mosquitos tend to cluster. This approach — known as targeted indoor residual spraying — is too expensive and time-consuming to apply across a city. 

The statistical framework in the current paper, however, allows public health officials to concentrate their efforts on previous hot spots for Aedes-borne diseases to better control — and even prevent — outbreaks. 

“The ultimate goal is to give public health officials the power to harness big data and do more effective and efficient mosquito control — even before an epidemic begins,” Vazquez-Prokopec says. 

Vazquez-Prokopec is currently leading a consortium in a randomized clinical trial in Merida to test targeted indoor residual spraying as an intervention against Aedes-borne diseases. The five-year trial, launched in 2020, is funded by a $6.5 million grant from the National Institutes of Health. 

Dengue fever is sometimes called “break-bone fever” due to the excruciating pain that is among its symptoms. More than one third of the world’s population lives in areas at high risk for infection with the dengue virus, a leading cause of illness and death in the tropics and subtropics. Dengue is endemic through most of Mexico, where between 75,000 and 355,000 cases occur annually, translating into an economic cost of between about U.S. $150 million and $257 million annually. 

Chikungunya is rarely fatal but the symptoms can be severe and debilitating. Zika can cause symptoms similar to those of dengue and chikungunya, such as joint pains and fever. While Zika tends to be less debilitating, or even asymptomatic, if a pregnant woman contracts the virus it can have a catastrophic impact on her unborn child, including severe brain defects. 

Co-authors of the Lancet Planetary Health paper include researchers from Mexico’s Ministry of Health; the Autonomous University of the Yucatan; the Autonomous University of Nuevo Leon; Yucatan Health Services; the Mexican Society of Public Health; the National Institute of Public Health in Cuernavaca; the Undersecretary of Prevention and Health Promotion in Mexico City; the Pan American Health Organization; and the U.S. Centers for Disease Control and Prevention. 

The work was funded by USAID, the U.S. Centers for Disease Control and Prevention, the Canadian Institute of Public Health, the state of Yucatan, the National Institutes of Health and Emory University.

Related:

Zeroing in on a mosquito menace

Contact tracing, with indoor spraying, can curb dengue outbreak