Suzanne Bakken

Ali Zolnour, Hossein Azadmaleki, Yasaman Haghbin et al.

Alzheimer's disease and related dementias(ADRD) affect nearly five million older adults in the United States, yet more than half remain undiagnosed. Speech-based natural language processing(NLP) offers a scalable approach for detecting early cognitive decline through subtle linguistic markers that may precede clinical diagnosis. This study develops and evaluates a speech-based screening pipeline integrating transformer embeddings with handcrafted linguistic features, synthetic augmentation using large language models(LLMs), and benchmarking of unimodal and multimodal classifiers. External validation assessed generalizability to a MCI-only cohort. Transcripts were drawn from the ADReSSo 2021 benchmark dataset(n=237, Pitt Corpus) and the DementiaBank Delaware corpus(n=205, MCI vs. controls). Ten transformer models were tested under three fine-tuning strategies. A late-fusion model combined embeddings from the top transformer with 110 linguistic features. Five LLMs(LLaMA8B/70B, MedAlpaca7B, Ministral8B,GPT-4o) generated label-conditioned synthetic speech for augmentation, and three multimodal LLMs(GPT-4o,Qwen-Omni,Phi-4) were evaluated in zero-shot and fine-tuned modes. On ADReSSo, the fusion model achieved F1=83.3(AUC=89.5), outperforming transformer-only and linguistic baselines. MedAlpaca7B augmentation(2x) improved F1=85.7, though larger scales reduced gains. Fine-tuning boosted unimodal LLMs(MedAlpaca7B F1=47.7=>78.7), while multimodal models performed lower (Phi-4=71.6;GPT-4o=67.6). On Delaware, the fusion plus 1x MedAlpaca7B model achieved F1=72.8(AUC=69.6). Integrating transformer and linguistic features enhances ADRD detection. LLM-based augmentation improves data efficiency but yields diminishing returns, while current multimodal models remain limited. Validation on an independent MCI cohort supports the pipeline's potential for scalable, clinically relevant early screening.

Lana Yarosh, Suzanne Bakken, Alan Borning et al.

Substance Use Disorders (SUDs) involve the misuse of any or several of a wide array of substances, such as alcohol, opioids, marijuana, and methamphetamine. SUDs are characterized by an inability to decrease use despite severe social, economic, and health-related consequences to the individual. A 2017 national survey identified that 1 in 12 US adults have or have had a substance use disorder. The National Institute on Drug Abuse estimates that SUDs relating to alcohol, prescription opioids, and illicit drug use cost the United States over $520 billion annually due to crime, lost work productivity, and health care expenses. Most recently, the US Department of Health and Human Services has declared the national opioid crisis a public health emergency to address the growing number of opioid overdose deaths in the United States. In this interdisciplinary workshop, we explored how computational support - digital systems, algorithms, and sociotechnical approaches (which consider how technology and people interact as complex systems) - may enhance and enable innovative interventions for prevention, detection, treatment, and long-term recovery from SUDs. The Computing Community Consortium (CCC) sponsored a two-day workshop titled "Computational Support for Substance Use Disorder Prevention, Detection, Treatment, and Recovery" on November 14-15, 2019 in Washington, DC. As outcomes from this visioning process, we identified three broad opportunity areas for computational support in the SUD context: 1. Detecting and mitigating risk of SUD relapse, 2. Establishing and empowering social support networks, and 3. Collecting and sharing data meaningfully across ecologies of formal and informal care.