Research Publications

January 2026

Bahr-Hosseini M, Asghariahmadabad M, Bikson M, Saver JL, Liebeskind DS, Nael K.

In the first-in-human proof-of-concept TESSERACT study (Transcranial Electrical Stimulation in Stroke Early After Onset Clinical Trial), delivering high-definition cathodal transcranial direct current stimulation (HD C-tDCS) to penumbral tissue was shown to be a feasible and tolerable treatment strategy for acute ischemic stroke.¹ Additionally, in this dose-escalation study, signals of efficacy were observed: patients receiving active HD C-tDCS had a higher proportion of penumbral salvage (median, 66%) compared with sham patients (median, 0%) on standard hemodynamic perfusion imaging. The mechanisms through which HD C-tDCS may salvage penumbra are collateral enhancement via electrical current-induced pial arterial vasodilation and direct cytoprotection via inhibition of peri-infarct excitotoxicity.²

January 2026

Lookstein RA, Konstantinides SV, Weinberg I, Dohad SY, Rosol Z, Kopeć G, Moriarty JM, Parikh SA, Holden A, Channick RN, McDonald B, Nagarsheth KH, Yamada K, Rosovsky RP; STORM-PE Trial Investigators.

BACKGROUND: Patients with intermediate-high risk pulmonary embolism (PE) have an elevated right ventricular (RV) to left ventricular (LV) diameter ratio and are at risk of early clinical decompensation and mortality. Reperfusion therapy aims to rapidly relieve acute RV pressure overload and to normalize hemodynamics. STORM-PE (A Prospective, Multicenter, Randomized Controlled Trial Evaluating Anticoagulation Alone Versus Anticoagulation Plus Mechanical Aspiration With the Indigo Aspiration System for the Treatment of Intermediate-High Risk Acute Pulmonary Embolism) is the first reported randomized controlled trial to test the efficacy and to evaluate the safety of mechanical thrombectomy, specifically computer-assisted vacuum thrombectomy (CAVT) with anticoagulation compared to anticoagulation alone. METHODS: STORM-PE is an international randomized controlled trial with 1:1 randomization to CAVT with anticoagulation or anticoagulation alone. Eligible adults had acute-onset (symptoms ≤ 14 days), intermediate-high risk PE and were normotensive, with an RV/LV ratio ≥ 1.0 on computed tomographic pulmonary angiography and elevated cardiac biomarkers. The primary end-point analysis tested for a difference between groups for the change in RV/LV ratio at 48 hours, assessed by a blinded independent imaging core laboratory. Secondary end points included major adverse events within 7 days (a composite of clinical deterioration necessitating rescue therapy, PE-related mortality, symptomatic recurrent PE, and major bleeding), adjudicated by an external clinical events committee. Additional outcomes included change in vital signs and core laboratory-assessed pulmonary artery obstruction at 48 hours.

January 2026

May FP, McEwan JA, Tuan JJ, Crespi CM, Maehara CK, Yang JO, Uy A, Gupta S, Kang Y, Myint A, Naini BV, Elmore JG, Kadiyala S, Ong MK, Bui AAT, Hamilton AB, Chang LC, Gelvezon A, Hsu W, Bastani R.

INTRODUCTION: Colorectal cancer (CRC) is largely preventable through the removal of precancerous polyps from the colon and rectum. Polyp surveillance guidelines recommend that individuals with polyps categorized as high-risk neoplasia (HRN) undergo surveillance colonoscopy 3 years after HRN diagnosis. However, over half of patients fail to receive their surveillance colonoscopy within this recommended timeframe, highlighting the need for effective interventions to improve 3-year colonoscopy surveillance rates. METHODS: A hybrid type 1 effectiveness-implementation, 2-group cluster-randomized trial is being conducted at 30 primary health care clinics in a large, urban academic health center in Southern California. The study includes two arms: (1) a multilevel intervention arm in which a natural language processing (NLP) algorithm processes electronic health record (EHR) data to facilitate the identification of patients with HRN and providers and patients receive electronic notification when surveillance colonoscopy is due; and (2) a usual care arm, where no intervention is applied.

January 2026

Padia SA, Modi S, Wehrenberg-Klee E, Adamo RD, Ahmed A, Orloff M, Helmerhorst HJF, Chmielowski B, Ozkan OS, Burgmans MC, Grözinger G, Lewandowski RJ, Krishnasamy VP.

Uveal melanoma frequently metastasizes to the liver, with over 90% of metastatic cases involving hepatic spread. Despite systemic therapy advances, prognosis remains poor in hepatic-dominant disease. Percutaneous hepatic perfusion (PHP) with melphalan, now U.S. Food and Drug Administration (FDA)-approved via the Hepzato Kit, delivers high-dose chemotherapy directly to the liver while reducing systemic toxicity through extracorporeal filtration. This review outlines PHP's rationale, technique, patient selection, and institutional requirements. Ideal candidates have multifocal, bilobar liver metastases and limited extrahepatic disease. PHP requires coordinated care across interventional radiology, anesthesia, perfusion, and pharmacy. Clinical data show encouraging survival and disease control with manageable hematologic toxicity and rapid recovery. Research into improved filtration, optimized dosing, and immunotherapy integration may further improve outcomes.

January 2026

Roberts DG, Ramakrishnan A, Joglekar A, Sayre J, Haber Z, Srinivasa R.

BACKGROUND AND AIMS: Benign biliary strictures (BBSs) are associated with poor long-term patency, substantial morbidity, and high health-care costs. Despite a multitude of available treatment options-peroral endoscopic, percutaneous transhepatic, and surgical reconstruction-recurrence rates remain high. This study assesses the safety and efficacy of percutaneous transhepatic cholangioscopy (PTCS)-assisted holmium laser stricturotomy for BBS management. METHODS: The Percutaneous Endoscopic Benign Biliary Laser study is a prospective, single-arm, open-label pilot trial conducted at a single academic institution. Between October 2022 and August 2023, 5 patients with BBSs underwent PTCS-assisted laser stricturotomy, with or without lithotripsy, using a holmium:yttrium-aluminum-garnet laser fiber. Patients were followed for 12 months. Primary endpoints included technical success and 30-day adverse events, while secondary endpoints assessed 12-month primary patency (freedom from stricture recurrence) and cumulative device-free survival. Data were gathered through imaging (magnetic resonance cholangiopancreatography or CT), clinical visits, and laboratory tests at 3, 6, and 12 months.

January 2026

Shahrvini T, Wood EJ, Joines MM, Nguyen H, Hoyt AC, Chalfant JS, Capiro NM, Fischer CP, Sayre J, Hsu W, Milch HS.

BACKGROUND. Insights into the nature of false-positive findings flagged by contemporary mammography artificial intelligence (AI) systems could inform the potential use of AI to reduce false-positive recall rates. OBJECTIVE. The purpose of this study was to compare AI and radiologists in terms of characteristics of false-positive digital breast tomosynthesis (DBT) examinations in a breast cancer screening population. METHODS. This retrospective study included 2977 women (mean age, 55 years) who were participating in an observational population-based screening study and underwent 3183 screening DBT examinations from January 2013 to June 2017. A commercial AI tool analyzed DBT examinations. Positive examinations were defined as having an elevated-risk result for AI and as having been assigned BI-RADS category 0 for interpreting radiologists. False-positive examinations were defined as the absence of a breast cancer diagnosis within 1 year. Radiologists rereviewed the imaging for AI-flagged false-positive findings.

January 2026

Welland SH, Kim GHJ, Yadav A, Nael K, Hoffman JM, Brown MS, McNitt-Gray MF, Hsu W.

PURPOSE: There are multiple commercially available, Food and Drug Administration (FDA)-cleared, artificial intelligence (AI)-based tools automating stroke evaluation in noncontrast computed tomography (NCCT). This study assessed the impact of variations in reconstruction kernel and slice thickness on two outputs of such a system: hypodense volume and Alberta Stroke Program Early CT Score (ASPECTS). APPROACH: The NCCT series image data of 67 patients imaged with a CT stroke protocol were reconstructed with four kernels (H10s-smooth, H40s-medium, H60s-sharp, and H70h-very sharp) and three slice thicknesses (1.5, 3.0, and 5.0 mm) to create 1 reference condition (H40s/5.0 mm) and 11 nonreference conditions. The 12 reconstructions per patient were processed with a commercially available FDA-cleared software package that yields total hypodense volume (mL) and ASPECTS. A mixed-effect model was used to test the difference in hypodense volume, and an ordered logistic model was used to test the difference in e-ASPECTS.

January 2026

Zhang W, Kafali SG, Adamos T, Kuwahara K, Dong A, Li J, Shih SF, Delgado-Esbenshade T, Chowdhury S, Loong S, Moretz J, Barnes SR, Li Z, Ghahremani S, Calkins KL, Wu HH.

BACKGROUND: Pediatric abdominal visceral and subcutaneous adipose tissue (VAT, SAT) quantified on magnetic resonance imaging (MRI) can assess risk for metabolic diseases. However, the complex structure of VAT in children and the lack of sufficient MRI datasets pose challenges for developing automated segmentation methods. PURPOSE: To achieve accurate and rapid automated segmentation of pediatric abdominal VAT and SAT on motion-robust free-breathing (FB) 3D Dixon MRI by developing a cross-cohort federated learning (FL) framework that leverages adult datasets. MATERIALS AND METHODS: 3D FB-MRI datasets were prospectively acquired in children 6-18 years old (single center, 2 scanners; 2016-2023) and used to train 3D neural network models for segmenting abdominal VAT and SAT. The FL model was trained across the pediatric cohort and a separate adult cohort (5 centers, 7 scanners) without requiring direct data sharing. Segmentation performance of the FL model was assessed by Dice scores with respect to references and compared with standalone local training and joint training with full data access. Quantification of VAT and SAT volume and proton-density fat fraction (PDFF) was compared against references using intraclass correlation coefficients (ICCs) and Bland-Altman analysis. Differences between training approaches were analyzed using the Kruskal-Wallis test followed by paired Wilcoxon signed-rank tests.