Treatment Optimization in Healthcare: Director of Data Science and AI at AccessHope on AI Innovation

In oncology, where every decision can change a patient’s life, AI is empowering clinicians with precise treatment choices and proactive care strategies, turning data into hope. In an exclusive interview, Ken Kwasniewski, Director of Data Science and AI at AccessHope, shares insights on how AI transforms healthcare through treatment regimen optimization and anticipatory care coordination. He tackles challenges like cultural misconceptions, workflow alignment, and data integration, offering solutions such as explainable tools and ETL pipelines. Ken highlights benefits, including enhanced oncology outcomes, and envisions AI as a real-time care GPS. This discussion traces AI’s journey in healthcare, offering actionable insights for clinicians and health systems seeking sustainable, predictive innovation.

Introduction: The Rise of AI in Healthcare

Healthcare grapples with treatment variability, fragmented care coordination, and rising costs that burden providers and patients. AI offers precision, optimizing regimens and predicting needs to enhance outcomes. Ken Kwasniewski, Director of Data Science and AI at AccessHope, leverages his OncoHealth experience to advance AI-driven solutions. His work focuses on tools that improve oncology care and align with clinical goals. This blog follows AI’s evolution in healthcare, from today’s treatment optimization to future dynamic care models, detailing Ken’s strategies for overcoming barriers like clinician resistance and data complexity, paving the way for patient-centered innovation.

Current State: Optimizing Treatment Regimens

AI is reshaping oncology by optimizing treatment regimens, aligning oncologist decisions with outcomes data to deliver precise care. This approach reduces unwarranted variation, ensuring consistent, effective treatments. However, variability in provider decisions poses a challenge, as differing practices can lead to suboptimal outcomes and higher costs.

Ken Kwasniewski, Director of Data Science and AI at AccessHope, noted that AI creates “immediate value in oncology care.” His team uses tools like TRACE to align regimens with data-driven insights, enhancing provider accountability.

• Aligns decisions with outcomes data for precision.

• Enhances accountability among providers.

• Improves clinical consensus for better care.

Ken explained that TRACE’s impact stems from its focus on real-world utility, drawing from his OncoHealth experience. At AccessHope, this approach ensures clinicians deliver consistent, evidence-based care, improving patient outcomes. The result is measurable: enhanced oncology care and reduced variation, solidifying AI’s role in modern healthcare.

Overcoming Cultural Barriers

Scaling AI in healthcare faces cultural resistance, with misconceptions that “AI replaces clinicians” deterring adoption. This fear undermines trust, as providers worry about losing control over patient care, slowing the integration of predictive tools.

Ken highlighted the need to counter this myth. At AccessHope, his team builds explainable tools like TRACE that contextualize predictions and incorporate clinician feedback to foster trust.

• Contextualizes predictions for clinical relevance.

• Integrates feedback to refine AI models.

• Builds trust through transparent outputs.

He emphasized that AI should support, not supplant, clinicians, a principle guiding AccessHope’s strategy. By prioritizing explainability, Ken ensures tools align with provider needs, encouraging adoption. This approach delivers trusted insights, enabling clinicians to enhance care with confidence and countering cultural resistance.

Aligning AI with Workflows

Embedding AI into clinical and operational workflows is critical for impact, but ensuring outputs are relevant to daily tasks is challenging. Irrelevant predictions can disrupt workflows, reducing efficiency and limiting AI’s value in healthcare settings.

Ken explained that AccessHope approaches AI from a value-based lens. TRACE integrates with claims verification and triggers operational actions, aligning predictions with clinical and payer goals.

• Triggers actions like care planning or telehealth.

• Aligns with enterprise KPIs for efficiency.

• Enhances payer-clinician collaboration.

He noted that TRACE is a “strategic differentiator” at AccessHope, streamlining workflows and supporting value-based care. This integration ensures AI drives actionable outcomes, improving member satisfaction and reducing costs. By aligning with workflows, Ken’s strategy enhances AI’s practical impact in oncology.

Integrating Multi-Modal Data

Bridging current AI to predictive models requires integrating diverse data—EHRs, claims, genomics, and social determinants. Harmonizing these sources is a challenge, as their differing formats and languages complicate real-time predictions.

Ken described integration as a “layered translation challenge.” At AccessHope, his team uses ETL pipelines on Databricks and LLMs like LangChain to stitch structured and unstructured data for predictive insights.

• Stitches structured data with Databricks and PySpark.

• Translates narratives using LangChain LLMs.

• Aims for real-time synthesis of new data.

He highlighted the goal of a real-time synthesis engine at AccessHope, capable of updating recommendations dynamically. This approach supports anticipatory care by delivering holistic insights, enabling proactive interventions that enhance oncology outcomes.

Ensuring Governance and Explainability

AI adoption in healthcare demands governance and explainability, as opaque recommendations risk undermining clinician trust. Without transparency, providers may disregard AI insights, limiting their impact on patient care.

Ken stressed that “transparency” is the foundation of governance. At AccessHope, TRACE includes LLM-powered commentary layers and post-deployment audits to ensure accountability and trust.

• Provides confidence intervals and clinical evidence.

• Allows clinicians to interrogate insights.

• Reviews overrides to improve models.

He explained that maintaining trust requires ongoing accountability, a priority at AccessHope. By enabling clinicians to question predictions and learn from overrides, Ken’s framework fosters confidence. This governance approach ensures AI delivers credible, trusted insights, promoting adoption in clinical settings.

Future Horizons: Dynamic Patient Journeys

AI’s future in healthcare lies in dynamic patient journeys, acting as a real-time care GPS that adjusts based on clinical, behavioral, and environmental signals. Scaling such personalized coordination is challenging, requiring robust models to orchestrate care proactively.

Ken envisioned AI as a “real-time interpreter” of patient journeys. At AccessHope, his team develops predictive models for anticipatory care coordination, tailoring treatments and support ecosystems.

• Orchestrates actions across care teams proactively.

• Adjusts care based on real-time signals like lab results.

• Tailors support with telehealth and caregiver interventions.

He predicted that AI will redefine oncology by integrating diverse signals, a vision driving AccessHope’s focus. This approach promises personalized, resilient care, enhancing outcomes over the next five years. Ken’s strategy positions AI to transform healthcare delivery with precision.

Ken Kwasniewski, Director of Data Science and AI at AccessHope, shares a roadmap for AI-driven healthcare in this exclusive interview. By tackling cultural misconceptions, workflow alignment, and data complexity with solutions like explainable tools and ETL pipelines, he charts a path to dynamic care. His insights on oncology outcomes and future care GPS highlight his expertise. This discussion underscores AI’s potential to transform healthcare with accountability.

FAQ: Exploring AI in Healthcare

• Q: How does AI transform healthcare?

  • AI revolutionizes oncology care, per Ken. It optimizes treatments and coordinates care proactively. AI aligns treatment regimens with outcomes data to ensure precision. It orchestrates actions across care teams to prevent issues before they arise. This approach significantly improves patient outcomes in oncology care settings.

• Q: What benefits has AI delivered in healthcare?

  • AI enhances oncology outcomes, per Ken. It ensures consistent, effective care. AI reduces unwarranted variation in treatment regimens for better consistency. It boosts provider accountability by aligning decisions with data. This leads to improved member satisfaction and more effective oncology care delivery.

• Q: How is data quality ensured for AI models?

  • AI integrates reliable data, per Ken. Advanced pipelines ensure accuracy. AccessHope uses ETL pipelines with Databricks to stitch structured data seamlessly. LangChain LLMs translate unstructured clinical narratives for clarity. Clinician feedback validates insights to maintain high data quality for predictive models.

• Q: Who controls healthcare data in AI systems?

  • AccessHope prioritizes data governance, per Ken. Transparent processes ensure control. ETL pipelines securely integrate data while aligning with clinical and payer standards. Compliance frameworks protect sensitive information. This ensures healthcare data remains controlled and secure within AI-driven systems.

• Q: What challenges hinder AI adoption in healthcare?

  • Cultural and technical barriers persist, per Ken. Resistance and workflow issues slow progress. Clinicians fear AI automation may replace their roles. AI outputs must integrate seamlessly into workflows for relevance. Harmonizing multi-modal data like EHRs and genomics remains a significant technical challenge.

• Q: What’s the next AI breakthrough in healthcare?

  • AI will guide patient journeys, per Ken. Dynamic care coordination is the future. AI will proactively coordinate care across teams based on real-time signals. It will tailor support ecosystems, including telehealth and caregiver interventions. This approach integrates diverse health signals for personalized oncology care.

• Q: How does AI gain clinician trust?

  • AI fosters trust through transparency, per Ken. Explainable tools build credibility. TRACE provides confidence intervals and clinical evidence for each recommendation. Clinicians can interrogate insights to understand predictions. Post-deployment audits review overrides to continuously improve AI models and maintain trust.