NeoGenomics: Data in Oncology Testing & Diagnostics
The future of medicine and oncology specifically lies with data and algorithms. One of the ways that AI will impact the future is reducing disparities in patient care, especially in community settings, says Melody Harris, Chief Operations Officer & President of Informatics at NeoGenomics.
“We know that 80% to 85% of cancer care is happening in the community. Yet, those areas have a different level of access to testing or expertise in pathology,” she says. “How do we level-set clinics in smaller communities like Tulsa, Oklahoma, or Birmingham, Alabama, with those in New York City, Boston, or LA? Data!”
Melody likes to think of it as the digitisation of the body. By converting molecules taken from our bodies into binary data—ones and zeros—computers can generate additional insights.
“Leveraging datasets from patients who have undergone extensive testing, community doctors can utilise this information to adopt a precision medicine approach in selecting treatments tailored based on anticipated outcomes, thereby decreasing disparities,” she says.
Data and AI in oncology, automation and machine learning in laboratory processes
Melody’s interest in health-related data began when she was diagnosed with cancer almost 20 years ago. During the diagnostic phase, she was curious and hopeful about how data from her own body and tumour could help guide the decisions needed around treatment.
“The test didn’t give me the precision medicine I was hoping for, but the experience led me to want to use data to pursue true precision medicine,” she explains.
What followed were a series of positions at HealthyCircles, Qualcomm Life and SomaLogic, each leading her closer to making a meaningful change in people’s cancer journey, using insights derived from personalised data.
“My current position at NeoGenomics allows me to explore what can be done with millions of data points and create meaningful insights for oncology patients and providers,” she says. “Then, I work with my team to make those possibilities a reality, aiming to improve patient outcomes.”
Automation and machine learning are changing laboratory processes by enhancing efficiency, accuracy and the ability to manage complex data. Routine tasks such as sample preparation and data entry are streamlined significantly increasing throughput and quality. This leads to more reliable results and the ability to handle larger volumes of tests with consistent precision.
“Software automation and machine learning complement these advancements by digitising the lab and automating processes rather than completing them manually,” she continues. “Machine learning algorithms excel in identifying patterns and correlations in large datasets, a crucial advantage in oncology testing, where genetic and molecular data are voluminous and complex.”
This capability speeds up analysis and improves the predictive accuracy of tests, which is vital for early detection and personalised treatment strategies.
Data benefits biopharma providers with therapeutic drug development and selection
Data is crucial in therapeutic drug development and selection, from initial discovery to commercialisation. At NeoGenomics, the team’s role is to provide a holistic from initial discovery to commercialisation data-driven approach that supports biopharma partners through every stage of the product life cycle, from early discovery to market launch.
“We collaborate with our pharma partners at the beginning stages to focus on early assay design and development. This phase involves researchers and scientists defining their drug's potential impact on patient care and conducting preliminary clinical trials,” Melody says. “Our data capabilities become increasingly important as the drug progresses through clinical trials and into commercialisation. By studying the patient journey, we can help generate and analyse top-line data to forecast market potential and effectively position the drug. We look at factors such as treatment efficacy, treatment pathways, standards of care.”
The data also integrates information on genomic factors, co-expression and mutations, which are vital for optimising therapeutic strategies and making informed decisions about drug selection.
“By linking and integrating these diverse data sets, we support pharma partners in refining their drug's market strategy and ensuring successful commercialisation,” she says.
However, it is said that some of the leading players in health AI lack the data needed for robust insights. But Melody argues that it’s not that they lack any data, but they don’t have the breadth of rich, relevant data needed to power AI effectively.
“Companies can build all the required infrastructure, but they can’t create data—they either have to purchase it from aggregators or licence it from source data generators like NeoGenomics,” she explains. “At NeoGenomics, our data is abundant because of our expansive oncology-focused testing menu and our reach in the oncology space beyond large academic research centres. We generate large quantities of raw, real-world data for AI to learn on.”
The more data you have, the more you can feed an algorithm, which gets richer and more valuable with each influx of new data. It’s been said that data is the new oil—it will power the entire economy. The team at NeoGenomics cannot agree more, especially in oncology.
******
Make sure you check out the latest industry news and insights at Healthcare Digital and also sign up to our global conference series - Tech & AI LIVE 2024
******
Healthcare Digital is a BizClik brand