SEARCH · PubMed Health

Results for “multimodality”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources

PubMedDate not supplied

Radiotherapy-centered multimodal treatment of unresectable pancreatic carcinoma.

Multimodal treatment procedures, including intraoperative and external beam radiotherapy, chemotherapy and hyperthermotherapy, used for treatment of unresectable pancreatic carcinoma for the past two years have been described. Among the ten progressive cases where multimodal treatment was applied, marked reduction in tumor mass was observed in three cases. The cases receiving such treatment reported prolonged survival, the median survival period being 250 days as compared with 85.4 days in the non-multimodal group. The conclusion is that optimal palliative effects can be achieved by sustained application of both intraoperative and postoperative radiotherapy, hyperthermia and other techniques of multimodal therapy in cases of unresectable pancreatic carcinoma.

Adult↗

PubMed1990

[Goals, results and limitations of multimodal tumor therapy].

Multimodal tumor therapy, or the employment of two or more treatment modalities in combination, had led to several advances in the cure or long-term survival of patients with various tumor types. The individual established indications for multimodal tumor therapy are listed and described. Other indications remain controversial or are under evaluation in prospective randomized studies. In certain tumor types no advantage from the use of multimodal treatment has been demonstrated as yet. Aside from the advantages of multimodal therapy, the problems involved and possible side effects, especially in regard to short-, mid- and long-term toxic effects are described.

Antineoplastic Combined Chemotherapy Protocols↗

PubMed1989

[Multimodality treatment of carcinoma of the pancreas].

Although surgical resection has been the mainstream treatment for carcinoma of the pancreas, the operative results have been so disappointing that most surgeons in western countries have given up performing the resectional procedure. On the contrary, Japanese surgeons have never abandoned their dream of surgical treatment as a cure for the disease. Therefore, more and more aggressive procedures have been performed. Our operative results have not so remarkably ameliorated, but we have become knowledgeable on the pathological features of the carcinoma and believe that the best procedure for carcinoma of the head of the pancreas is a pancreatoduodenectomy with extensive dissection of regional lymph nodes and retroperitoneal tissue, and that surgery itself can not cure the disease but multimodality treatment should be established. Two hundred cases with carcinoma of the pancreas in which cystadenocarcinoma and islet cell carcinoma were excluded, were encountered from 1969 to 1987 in our department. Of 200 cases, only 48 cases underwent resection. Resection was divided into curative and non-curative resection according to macroscopic findings and pathohistological examination of the resected specimen. In cases of curative resection group, the average survival period of cases which underwent multimodality treatment was much longer than that without any adjuvant treatment. However, in cases of noncurative resection group, average survival period of cases with multimodality treatment was almost the same as that without adjuvant therapy. Therefore, multimodality treatment should be applied for curatively resected cases in order to obtain better results. Radiation therapy, especially intraoperative radiation therapy is considered to be a promising alternative modality of extensive retroperitoneal dissection. Hepatic metastasis was found postoperatively in about 27 percent of the resected cases. It seems that this type of recurrence occurred by migration of malignant cells from the tumor into the portal vein due to operative manipulation during surgery. Therefore, intraoperative infusion of an anticancer agent through the portal vein is mandatory, and preoperative and postoperative adjuvant chemotherapy should be considered.

Combined Modality Therapy↗

PubMed1988

[Multimodal treatment of cancer of the biliary tract].

The incidence of cancer of the biliary tract has been recently increasing, but the results of treatment have been unsatisfactory. During the last 10 years and 10 months, 128 cases of carcinoma of the biliary tract, including 64 cases of the gallbladder and the bile duct, respectively, were admitted. Some 98 (86%) out of the 113 cases were resected, with a low curative surgery rate of 32.7%. The curative surgery even resulted in recurrence with a few long-term survivors, so multimodal treatment should be considered for all cases. In non-curative resection, the 2-year cumulative survival rate of gallbladder carcinoma was 25% with radiation and chemotherapy, compared to the group without such treatment, all of whom died within 2 years after surgery. In cancer of the bile duct, similar results were obtained, so multimodal treatment should be administered especially in non-curative resection cases. In 1985 Mizumoto's group reviewed 1614 cases of gallbladder carcinoma and bile duct carcinoma collected from 22 institutions in Japan. The resectability and curative rate have been increasing, and the survival rates of both groups have slightly increased. Multimodal treatment has involved radiation therapy in 13.6% and chemotherapy in 44.1% of the cases. A two-year cumulative survival rate increased in non-curative resection patients treated with multimodal treatment.

Bile Duct Neoplasms↗

PubMed1985

Multimodal frequency distribution analysis of peripheral nerves.

In morphometric studies of peripheral nerves, the statistical analysis of such data as axon diameters is complicated by the presence of multimodal distributions. Nerve fiber diameters, for example, cannot be analyzed by classical parametric tests, and such descriptive statistics as mean or variance lose much of their usefulness. The recent development of stochastic catastrophe models offers a new parametric tool with which to describe multimodal distributions. This paper describes our development of a PASCAL computer program that permitted the modelling, comparison and segmentation of multimodal distributions. The method is based on a description of the multimodal frequency distributions by probability density functions of the canonical exponential family.

Axons↗

PubMed2026

Data-centric, robust, and explainable multimodal deep learning for clinical decision support: A systematic review.

PURPOSE: Multimodal deep learning is increasingly proposed for clinical decision support (CDS) under a "data-centric" framing that prioritizes label quality, missing-modality robustness, distribution shift, calibration, and explainability. Prior reviews have examined multimodal medical AI, CDS, and data-centric methods separately, but none address their intersection. We mapped the modalities, fusion strategies, and data-centric and explainability techniques used in this recent literature, quantified how often each is implemented rather than merely mentioned, assessed deployment-relevant evidence (external validation, clinical-outcome measurement, equity), and formally appraised study-level risk of bias. METHODS: Following the PRISMA 2020 statement (PROSPERO CRD420261427815; registered retrospectively), we screened 150 records and included primary, clinical, multimodal studies that applied machine or deep learning to a decision-support task and reported at least one quantitative result. Two reviewers screened and extracted data with consensus adjudication. Each study was coded against pre-specified operational definitions, separating implemented or empirically evaluated techniques from those only mentioned. Study-level risk of bias was assessed with PROBAST + AI. Synthesis was narrative. RESULTS: Thirty-one studies met inclusion; 30 (97%) were published between 2024 and 2026, with a median of three modalities (range 2-6), most commonly structured EHR (71%) and imaging (39%). Data-centric techniques were frequently reported (74-84% across label-noise, distribution-shift, calibration, missing-modality and class-imbalance handling; equity 61%). However, external validation was reported in only 4/31 studies (13%), a clinical or provider outcome in 3/31 (10%), and no study reported routine deployment. Overall risk of bias was high in 27/31 studies (87%), driven by the analysis domain. CONCLUSION: Within this recent, self-selected slice of the field, technical robustness and explainability techniques are widely reported but rarely validated out-of-distribution or against clinical outcomes, and the underlying evidence is at high risk of bias. Progress requires external multi-site validation, clinical-outcome measurement, formal bias appraisal, and adherence to AI reporting standards (e.g., TRIPOD + AI) before deployment can be justified.

Deep Learning↗

PubMed2026

Impact of a multimodal prehabilitation program on postoperative cognitive dysfunction: a single-center randomized controlled trial.

BACKGROUND: Postoperative cognitive dysfunction (POCD) is a frequent complication after cardiac surgery. Exercise-based prehabilitation may enhance functional reserve and reduce vulnerability to perioperative cerebral insults. We hypothesized that multimodal prehabilitation reduces POCD 3 months after cardiac surgery. METHODS: This prespecified substudy of a single-center randomized controlled trial (NCT03466606) included patients aged ≥50 years undergoing elective coronary artery bypass grafting and/or valve surgery. Participants were randomized 1:1 to 4-6 weeks of multimodal prehabilitation (exercise training, nutritional support, and psychological support) or standard preoperative care. Cognitive function was assessed at baseline and 3 months postoperatively using an age- and education-adjusted neuropsychological battery. POCD was defined as performance ≥1.5 standard deviations below normative values in at least 2 cognitive tests, excluding the Mini-Mental State Examination. Logistic regression analyses were performed to evaluate factors associated with POCD. RESULTS: Of 160 participants screened from the parent trial, 134 met eligibility criteria for the substudy and were randomized; 116 completed 3-month follow-up (prehabilitation n = 53; control n = 63). POCD occurred in 29 patients (25%), including 15/53 (28%) in the prehabilitation group and 14/63 (22%) in controls (odds ratio [OR] 1.37, 95% confidence interval [CI] 0.54-3.50, P = 0.52). In multivariable analysis, preoperative cognitive impairment was independently associated with POCD (OR 13.28, 95% CI 4.06-43.41, P < 0.001), whereas prehabilitation was not (OR 1.09, 95% CI 0.35-3.45, P = 0.877). Higher physical activity levels at 3 months were associated with lower odds of POCD (OR 0.97, 95% CI 0.95-1.00, P = 0.047). CONCLUSIONS: In this randomized controlled trial, a 4-6-week multimodal prehabilitation program did not reduce postoperative cognitive dysfunction 3 months after cardiac surgery. Although the intervention did not achieve measurable cognitive protection, the observed association between postoperative physical activity levels and postoperative cognitive dysfunction warrants further investigation.

Humans↗

PubMed2025

Foundation model based multimodal transformer framework for survival analysis in HER2 stratified breast cancer.

Objective. To improve survival prediction for HER2-positive breast cancer by integrating histopathological, molecular, and clinical data using a multimodal transformer framework.Approach. We propose a multimodal transformer framework for breast cancer survival prediction using HER2 stratified (SurvMBC), a foundation model-enhanced architecture that fuses three data modalities: whole-slide images, clinical narratives, and molecular features. Tumor microenvironment features are extracted using a pathology language and image pre-training (PLIP), clinical narratives are processed with BioBERT, and miRNA expression plus DNA methylation data are embedded using Gen2Vec. These representations are integrated through a cross-modal transformer with attention mechanisms for survival prediction.Main results. The model was evaluated on 1,095 HER2-positive breast cancer patients from The Cancer Genome Atlas. SurvMBC achieved a concordance index (C-index) of 0.857 (95% CI: 0.834, 0.880), a low integrated Brier score, and a strong inverse negative binomial log-likelihood. Risk stratification based on model outputs significantly separated high- and low-risk groups (log-rankp< 0.01) and showed strong associations with tumor stage, grade, and hormone receptor status (allp< 0.05).Significance. SurvMBC demonstrates the effectiveness of multimodal fusion in addressing tumor heterogeneity and improving prognostic accuracy. The attention-based integration enables context-aware learning of survival-relevant features across modalities, supporting individualized risk stratification and risk-adaptive treatment planning for HER2 stratified breast cancer patients.

Breast Neoplasms↗

PubMed2025

Multimodal deep learning for immunotherapy response prediction and biomarker discovery in non-small cell lung cancer.

OBJECTIVE: Immunotherapy has emerged as a promising treatment for advanced non-small cell lung cancer (NSCLC), but accurately predicting which patients will benefit from it remains a major clinical challenge. To address this, we aim to develop a novel multimodal method, DeepAFM, that integrates histopathology, genomic features, and clinical information to predict patient responses to anti-PD-(L)1 immunotherapy. MATERIALS AND METHODS: A total of 93 patients with advanced NSCLC were included in this study. Histopathological whole-slide images were processed using a self-supervised VQVAE2 for representation learning. PCA and K-means clustering were then applied for dimensionality reduction and feature grouping. Key regions of interest were visualized through permutation importance evaluation and color-coding techniques. The extracted histopathological features, along with genomic alterations and clinical variables, were integrated into the DeepAFM multimodal prediction model. RESULTS: The DeepAFM achieved a high predictive performance with an area under the curve (AUC) of 0.77 (95% confidence interval: 0.69-1.00). Attention-based heatmaps revealed that the model could identify critical pathological patterns, genomic mutations, and clinical indicators associated with patient responses to immunotherapy. DISCUSSION: The integration of multimodal data enabled the model to capture complex interactions among pathology, genomics, and clinical characteristics, enhancing the interpretability and predictive power of immunotherapy response prediction. The visualization techniques facilitated the identification of biologically meaningful features and potential biomarkers. CONCLUSION: This study demonstrates the effectiveness of the DeepAFM in predicting responses to immunotherapy in advanced NSCLC. The approach not only improves prediction accuracy but also provides valuable insights for personalized treatment strategies and biomarker discovery.

Humans↗

PubMed2026

Deep learning-based multimodal pathogenomics integration for precision cancer prognosis.

BACKGROUND: Recent studies have revealed valuable prognostic insights in haematoxylin and eosin (H&E)-stained histological sections and transcriptomic profiles, suggesting potential applications in machine learning. However, existing methods lack sufficient intra- and inter-modal interactions, and face challenges in clinical validation due to incomplete multimodal data. METHODS: We proposed PathoGems (PathoGenomics-based integrative survival prediction), a weakly-supervised, interpretable multimodal learning framework that integrates histology and genomic profiles for precise cancer prognosis prediction. To evaluate the robustness of PathoGems, we initially curated a dataset of 1965 cases across four cohorts from The Cancer Genome Atlas (TCGA), including breast, colorectal, glioblastoma, and esophageal cancers. For external validation, PathoGems was further evaluated on four independent cohorts, consisting of 76 breast cancer and 41 esophageal squamous cell carcinoma cases from Zhejiang Cancer Hospital, as well as 102 colorectal cancer and 58 glioblastoma cases from the Clinical Proteomic Tumor Analysis Consortium (CPTAC). RESULTS: PathoGems effectively stratified patients into favorable and unfavorable risk groups, revealing significant differences in histological patterns, genomic features, and overall survival (log-rank test, p < 0.05). Moreover, the model’s predictions are further supported by visualization and transcriptomic analysis, enhancing interpretability and reliability. CONCLUSIONS: By fusing histological and clinicogenomic multimodal models, PathoGems will provide a solid foundation for developing an innovative tool that aids clinicians in making informed decisions and selection personalized treatment strategies for cancer patients.

Humans↗

PubMed2026

Multimodal Deep Learning and Foundation Models for Early Detection and Forecasting of Plant Diseases.

Plant diseases destroy 20-40% of global food production annually, posing a critical threat to food security for a projected population of 9.7 billion by 2050. Conventional diagnostic approaches relying on expert visual assessment are slow, costly, and unsuitable for modern agricultural scales. While deep convolutional neural networks demonstrated early promise, single-modality, image-centric systems consistently fail under real-world field conditions characterized by variable lighting, co-occurring infections, and cultivar diversity. This review synthesizes a decade of progress across four interconnected frontiers: the evolution of deep learning architectures for plant disease detection; the adaptation of foundation models including CLIP, SAM, and DINOv2 to agricultural contexts; the development of multimodal fusion frameworks integrating imagery, environmental, genomic, and hyperspectral data; and the transition from static disease diagnosis to descriptive comparison of reported metrics, which suggested that multimodal approaches frequently reported improved diagnostic performance relative to corresponding single-modality baselines, although direct cross-study comparison was limited by methodological heterogeneity. A systematic review following PRISMA guidelines identifies eligible comparative studies. Descriptive comparison of reported performance metrics across these studies indicated that multimodal approaches generally achieved higher accuracy and sensitivity than single-modality models, particularly for pre-symptomatic disease detection. Eight critical research gaps are identified, including the absence of a unified agricultural foundation model and limited climate-aware forecasting under non-stationary climate projections. A structured research agenda is proposed to accelerate translation from laboratory performance to globally equitable, field-deployable crop protection systems.

convolutional neural networks↗

PubMed1978

Multimodal immunotherapy of primary gastrointestinal tumors in rats. 1. Histologic correlation.

The effect of multimodal immunotherapy was studied in rats bearing primary gastrointestinal tumors induced by 1,2-dimethylhydrazine dihydrochloride. Multimodal immune manipulation consisted of combinations of splenectomy, C. parvum, unblocking serum, unblocked lymphoid cells, and levamisole. Such immunologic intervention resulted in significant inhibition of tumor growth, and their metastasis. Ten of 10 untreated rats, 8 of 8 rats treated with splenectomy alone and 10 of 10 rats treated with normal rabbit serum died of progressive tumor growth. None of the rats treated with combinations of splenectomy, unblocking serum, unblocked lymphoid cells, C. parvum and levamisole succumbed to progressive tumor growth during the observation period. The histologic evidence of tumor destruction was obtained in 18 of 22 tumors in rats of groups receiving multimodal immunotherapy.

Animals↗

PubMed2026

Multimodal features and prognostic risk assessment in locally advanced gastric cancer patients following neoadjuvant therapy based on machine learning algorithms: a multicenter study.

BACKGROUND: Neoadjuvant therapy (NAT) is recommended for locally advanced gastric cancer (LAGC), but some patients respond poorly. We aimed to construct a multimodal model integrating CT images, transcriptomic sequencing, and clinicopathological data to assess prognosis in LAGC patients receiving NAT. MATERIALS AND METHODS: This multicenter study included 505 LAGC patients who underwent NAT. Radiomic features were extracted from preoperative CT images of 505 patients. RNA-seq was performed on 277 post-NAT specimens, with additional data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases (n = 804). Patients were divided into training (168 cases), internal validation (72 cases), and external validation cohorts. Machine learning algorithms identified key radiomic, molecular, and clinical features associated with NAT response, which were then integrated into a multimodal model to predict overall survival (OS) and disease-free survival (DFS). RESULTS: Six radiomic and three molecular features significantly associated with NAT response were selected. Radiomic risk (hazard ratio [HR]: 4.0, P < 0.001) and molecular risk (HR: 7.1, P < 0.001) were independent prognostic factors. By integrating radiomic risk, molecular risk, and clinical characteristics, a multimodal model (MuMo) was constructed.The C-index results (OS, C-index = 0.855; DFS, C-index = 0.786) demonstrated that MuMo outperformed the single-modality models and ypTNM staging.Mechanistic analysis suggested that the efficacy of neoadjuvant therapy was significantly enriched in immune-inflammatory pathways. CONCLUSIONS: MuMo can effectively predict postoperative survival risk in LAGC patients receiving NAT, serving as a powerful tool for optimizing prognostic assessment.

Humans↗

PubMed1985

Stroop interference based on the multimodal correlates of haptic size and auditory pitch.

In a preliminary experiment subjects were asked to explore three wooden knobs of different sizes and to rate each one on a series of 7-point scales. The results confirmed that an object may possess a number of multimodal qualities that are contingent on its haptic size. The pattern of intercorrelations between the qualities was consistent with the pattern that is observed when subjects respond to pure auditory tones varying in pitch. For example, small (high-pitched) sounds, like small objects, are judged to be sharp, thin, light, weak, fast, tense, and bright. The main experiments used a paradigm based on the Stroop interference effect. Subjects were required to press one of two keys as quickly as possible depending on which of four possible words appeared in the centre of the screen. A 50 Hz or a 5500 Hz tone accompanied each test word, and subjects responded on two keys that differed in size. Subjects were found to respond more slowly when either the pitch of the incidental sound or the size of the key on which they responded was incongruent with the multimodal features represented by the test word. The results confirm that people are automatically and immediately sensitive to the multimodal features of a stimulus when direct sensory evidence for the features is absent.

Form Perception↗

PubMed2026

mmContext: an open framework for multimodal contrastive learning of omics and text data.

SUMMARY: Multimodal approaches are increasingly leveraged for integrating omics data with textual biological knowledge. Yet there is still no accessible, standardized framework that enables systematic comparison of omics representations with different text encoders within a unified workflow. We present mmContext, a lightweight and extensible multimodal embedding framework built on top of the open-source Sentence Transformers library. The software allows researchers to train or apply models that jointly embed omics and text data using any numeric representation stored in an AnnData.obsm layer and any text encoder available in Hugging Face. mmContext supports integration of diverse biological text sources and provides pipelines for training, evaluation, and data preparation. We train and evaluate models for a RNA-Seq and text integration task, and demonstrate their utility through zero-shot classification of cell types and diseases across four independent datasets. By releasing all models, datasets, and tutorials openly, mmContext enables reproducible and accessible multimodal learning for omics-text integration. AVAILABILITY AND IMPLEMENTATION: Pretrained checkpoints and full source code for our custom MMContextEncoder are available on Hugging Face huggingface.co/jo-mengr. The Python package github.com/mengerj/mmcontext provides the model implementation and training and evaluation scripts for custom training. The releases for the publication can be accessed via zenodo: adata_hf_datasets: doi.org/10.5281/zenodo.19185217 and mmContext: doi.org/10.5281/zenodo.19185493.

Computational Biology↗

PubMed1992

Ten-year outcome of patients with advanced epithelial ovarian carcinoma treated with cisplatin-based multimodality therapy.

PURPOSE: At the end of the 1970s it was thought that advanced epithelial ovarian cancer (EOC) could be cured by multimodality treatment using surgery, cisplatin-based combination chemotherapy, and radiotherapy (RT). Such multimodality treatment was used as standard therapy at our institution. Our long-term results are reviewed. PATIENTS AND METHODS: One hundred ninety-five previously untreated patients with stage III or IV EOC were treated between April 1979 and December 1982. All patients were to have debulking surgery, when feasible, followed by the administration of doxorubicin and cisplatin at 50 mg/m2 every 3 weeks until a total dose of doxorubicin of 450 mg/m2 had been reached. RT was used in addition in patients with disease remaining after the chemotherapy. Maintenance chemotherapy with oral cyclophosphamide and hexamethylmelamine (altretamine) was administered to patients who did not have a documented histologic complete remission. RESULTS: The 10-year overall and failure-free survivals were 4% and 8%, respectively. The median overall survival was 2 years. The achievement of a histologic complete response (n = 32) did not equate to cure because 20 (63%) of the patients eventually relapsed. Multivariate analysis identified residual disease of greater or less than 2 cm as the only independent prognostic factor. CONCLUSIONS: Our multimodality treatment program was noncurative for the majority of the patients. Innovative therapies are needed before we can hope to cure such disease.

Adult↗

PubMed1991

Multimodal radiography: a new imaging technique and system for oral diagnosis.

The multimodal radiography system is based on the combined use of narrow-beam radiography and spiral tomography in one set of equipment. The essentials of a system are a multifunction unit, preprogrammed imaging procedures for various dentomaxillofacial objects and a co-ordinate system for object location. Four years of clinical experience have shown that multimodal radiography is effective for depicting dentomaxillofacial structures for diagnostic purposes. The detailed narrow beam imaging mode of multimodal radiography has also been compared with intraoral radiography as regards their abilities to reveal periodontal and periapical lesions. Preliminary results indicate that detailed narrow beam radiography is more accurate than intraoral radiography in the diagnosis of periodontal and periapical pathology.

Equipment Design↗

PubMed1990

[Radical surgical intervention with conventional radiation versus multimodality therapy protocol in undifferentiated thyroid cancer].

Out of a total of 550 patients with thyroid cancer diagnosed over the 16-year period 1972-June 1989, 44 showed undifferentiated carcinoma and were treated by thyreoidectomy and early postoperative external irradiation. In order to analyse the outcome in patients treated by primary surgery in contrast to patients treated by means of a multimodal therapy concept we compared our surgical procedures with regard to primary surgical approach, early postoperative course, operative complications and survival to the data on the multimodal therapy concept of the Karolinska Hospital reported by E. Tallroth et al. 1987. A significantly better survival was correlated with radical (n = 20) versus palliative tumour resection (n = 24) (p less than 0.001), and total thyroidectomy (n = 25) versus subtotal thyroidectomy (n = 19) (p less than 0.006). Radical surgery with early postoperative external irradiation revealed no postoperative mortality and no symptomatic cervical tumour recurrence. By contrast, palliative surgery, particularly in the case of synchronous tracheotomy, was attended by a relatively high mortality (29%) and symptomatic local recurrences. The results of this study suggest that in undifferentiated thyroid carcinoma an attempt at radical tumour resection should be undertaken, since multimodal therapy procedures revealed a significantly highly complication rate (up to 36%) and, in comparison with a radical surgical treatment policy, showed a higher rate of local recurrences (0% vs. 48%) and a lower survival (mean survival 42 vs. 15 months).

Aged↗