SEARCH · Search PubMed

Results for “multimodal data”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

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

At least 235 records · Page 13Linked to original sources

PubMed2026

Advancing cancer detection and treatment using longitudinal routine clinical data.

Cancer management remains fragmented across its continuum, from late-stage diagnosis and salvage therapies to non-personalized surveillance. Here, we present Oncoformer, a unified multimodal transformer model trained on the China Oncology Multimodal Prediction and Surveillance Study (COMPASS) cohort (3.67 million individuals, 17.7 million clinical visits) and validated on independent external cohorts, including the UK Biobank. Oncoformer integrates longitudinal electronic health records with chest X-ray imaging to address multiple clinical tasks: pan-cancer diagnosis (area under the receiver operating characteristic curve [AUROC] = 0.956), future cancer prediction up to 1 year before diagnosis (AUROC = 0.869), tumor stage inference (mean AUROC > 0.90), patient-specific treatment-response forecasting, and recurrence-free survival stratification across ten cancer types (all p < 0.01). Staging predictions were independently validated against postoperative pathological endpoints and shown to converge on core cancer genomic pathways. By translating routine clinical data into a dynamic view of cancer evolution, Oncoformer provides a framework for risk-informed cancer prediction and treatment stratification using routine clinical data.

Humans↗

PubMed2004

Margin involvement and outcome in oesophageal carcinoma: a 10-year experience in a specialist unit.

AIMS: Positive microscopic margins after major cancer surgery adversely affect prognosis. We questioned whether the benefit of a multimodal approach in oesophageal carcinoma is due to reduced resection margin involvement and whether multimodal therapy alters the disease course when margins are involved. METHODS: Pathology specimens of 212 patients, treated with either multimodal therapy or surgery alone, were re-reviewed to assess margin involvement by tumour. Margin status was compared with recurrence and survival data. RESULTS: Margin involvement was decreased with multimodal therapy (16 of 103 patients) vs surgery (33 of 109 patients), associated with reduced tumour recurrence and a significant survival advantage. However, even with involved margins, multimodal therapy had lower recurrence vs surgery and a small survival benefit. CONCLUSIONS: Multimodal therapy significantly reduces margin involvement. The benefit of multimodal therapy remains highly significant for patients with clear margins. This study confirms for oesophageal carcinoma the value of an aggressive surgical approach in achieving negative resection margins.

Adenocarcinoma↗

PubMed2001

Iliofemoral deep vein thrombosis: conventional therapy versus lysis and percutaneous transluminal angioplasty and stenting.

OBJECTIVE: To compare conventional treatment (heparin and warfarin) of iliofemoral venous thrombosis with multimodality treatment (lysis and stenting). SUMMARY BACKGROUND DATA: Several studies have reported on conventional therapy for iliofemoral venous thrombosis with disappointing results. However, more recent studies have reported better results with multimodality treatment. METHODS: Fifty-one consecutive patients with extensive iliofemoral venous thrombosis were treated during a 10-year period. If there were no contraindications, patients were given the option to choose between conventional therapy (group 1) and multimodality therapy (group 2). The multimodality treatment strategy included catheter-directed lysis followed by percutaneous transluminal balloon angioplasty (PTA) and stenting for residual iliac stenoses. All patients underwent routine venous duplex imaging at 30 days, 3 months, 6 months, and every 6 months thereafter. RESULTS: There were 33 patients in group 1 and 18 patients in group 2. Demographic and clinical characteristics were comparable for both groups. Initial lysis was achieved in 16 of 18 patients (89%) in group 2. Ten of 18 patients in group 2 had residual stenosis after lysis (8 primary and 2 secondary to malignancy), and they were treated with PTA/stenting with an initial success rate of 90%. Two patients in group 1 (6%) had a symptomatic pulmonary embolism (none in group 2). At 30 days, venous patency and symptom resolution were achieved in 1 of 33 patients (3%) in group 1 versus 15 of 18 (83%) in group 2. Kaplan-Meier analysis showed primary iliofemoral venous patency rates at 1, 3, and 5 years of 24%, 18%, and 18% and 83%, 69%, and 69% for groups 1 and 2, respectively. Long-term symptom resolution was achieved in 10 of 33 patients (30%) in group 1 versus 14 of 18 (78%) in group 2. Kaplan-Meier life table analysis showed similar survival rates at 1, 3, and 5 years of 100%, 93%, and 85% for group 1 and 100%, 93%, and 81% for group 2. CONCLUSIONS: Lysis/stenting treatment was more effective than conventional treatment in patients with iliofemoral vein thrombosis.

Angioplasty, Balloon↗

PubMed1983

Intra-arterial adriamycin followed by surgery for limb sarcomas. Preliminary report.

A multimodal treatment schedule for operable soft tissue sarcomas is ongoing in our institution. The main purposes are to achieve an easier radical surgery with limb preservation and to lower the risk of local relapses. Patients with documented untreated soft tissue sarcoma of the lower limb are given continuous intra-arterial regional infusion with adriamycin consecutively for 8 days up to a dose of 100 mg/m2. Radical surgery is performed by the fourth day after infusion. Preliminary data on 13 cases are available: after infusion only 6 patients presented reduction of the tumor diameters, but almost all postoperative histological examinations revealed wide areas of necrosis (up to 100%) of tumor tissue (in 9 cases greater than 50%). In 4 patients surgical treatment was completed with radiotherapy. In 2 cases the preliminary indication to disarticulation could be modified in wide excision after the postinfusional tumor reduction. The toxicity of adriamycin was typical and acceptable. Further data will be available after adequate follow-up and when more patients enter the study. These data document the feasibility of this multimodal treatment and its validity in terms of specific tumor tissue toxicity and improved surgical approach.

Adolescent↗

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↗

PubMed1983

Multimodality evoked potentials and neurophysiological tests in multiple sclerosis. Effects of hyperthermia on test results.

Data on critical frequency of photic driving (CFPD), frequency following response (FFR), and visual, somatosensory (peroneal nerve), and brain-stem auditory evoked potentials (EPs) were obtained from 20 patients who had clinically definite multiple sclerosis and ten healthy normal subjects in a controlled, balanced study under normothermic and hyperthermic (+1 degrees C) conditions with a test-retest interval of one week. Normal subjects' test results showed no changes during hyperthermia. Patients' EP and CFPD data correlated well with history, clinical signs, and symptoms during both normothermia and hyperthermia. The FFR test data were equivocal and not fully analyzed. Data from the four other tests showed additional patient abnormalities during hyperthermia. Multimodality testing increased the number of patient abnormalities compared with single tests, and the number increased further during hyperthermia. Test-retest reproducibility was higher during hyperthermia.

Adult↗

PubMed2001

Multimodal cranial neuronavigation: direct integration of functional magnetic resonance imaging and positron emission tomography data: technical note.

OBJECTIVE: This is the first report of the direct integration of functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) data into cranial neuronavigation. METHODS: In a patient with a left precentral oligodendroglioma (World Health Organization Grade III), the Zeiss MKM system (Carl Zeiss Co., Oberkochen, Germany) was used for navigation based on thin-slice, T1-weighted, contrast-enhanced magnetic resonance imaging (MRI) scans. fMRI and methionine PET data were integrated by landmark matching, with reference to skin fiducials. RESULTS: The inaccuracy of the image fusion between fMRI and T1-weighted MRI data was 1.7 mm, that between PET and T1-weighted MRI data was 4.3 mm, and that for the subsequent registration of the navigation was 1.2 mm. The correct fMRI localization of the precentral gyrus was intraoperatively verified by cortical somatosensory evoked potential (phase-reversal) monitoring. Although the tumor was not clearly defined in the MRI scans, [11C]methionine PET demonstrated a clear tumor border, enabling us to achieve gross total tumor removal without postoperative functional deficits. CONCLUSION: Functional neuronavigation permits observation and preservation of relevant brain areas. Other functional areas (such as short-term memory areas) that can be detected only by fMRI might also warrant future monitoring. The simultaneous integration of fMRI and PET data adds a new dimension to cranial neuronavigation, enabling the observation of tumors in relation to functional cortical areas (in our case, the motor strip).

Brain↗

PubMed2001

Exploring extra-binomial variation in teratology data using continuous mixtures.

Discrete data from animal teratology experiments are known to exhibit extra-binomial variation. For example, we discuss a dominant lethal assay experiment in which male mice are exposed to various levels of radiation and are then mated to females. The response of interest is the number of resorptions out of the number of implantations. Most statistical work on analyzing such data has focused on modeling response rates as a function of dose of a suspected teratogen (radiation in this case) while accounting for the extra-binomial variability when calculating standard errors of the regression coefficients. Sometimes, however, when an unobserved genetic or exposure variable is suspected, the shape of the mixing distribution is of interest. We propose a mixture of beta-binomials (MBB) family of distributions that includes the non-parametric mixture of binomials model of Laird (1978) as a special case. The MBB family can accommodate a mixing distribution with one or more modes, and we develop a bootstrap test for multimodality. We apply the method to data from a dominant lethal teratology experiment.

Animals↗

PubMed1992

Central projections of axons from taste hairs on the labellum and tarsi of the blowfly, Phormia regina Meigen.

Taste hairs are located on the labellum and tarsi of blowflies. These multimodal hairs consist of four functionally distinct chemoreceptors and a mechanoreceptor. By staining selected multimodal hairs, we sought to identify the central projection patterns of multiple and single axons from those hairs. On each side of the labellum there are 11 "largest" hairs (LH). The neurons associated with the anteriormost (LH-1), posteriormost (LH-11), and one lateral (LH-6) hair on the labellum were stained selectively with cobaltous sulfide. The overall projection pattern in the central nervous system (CNS) for axons from LH-1 and LH-11 is similar and differs markedly from axons from LH-6. At least three individual axon-projection patterns were determined for each labellar hair filled, indicating a partial functional organization for axons from multimodal hairs. One identified axon, the dorsalmost axon, has terminal arborizations that do not differ with the location of its associated hair. Another axon, thicker than the others, projects to a region that is distinct from the four thin axons. Within this region the arborizations of the thick axons occupy different areas depending on the location of their associated hair. Neurons from the largest hairs on the distalmost tarsomere (D5) of each leg were also stained and consisted of one thick and four thin axons. All axons except one thin axon from tarsal D5 hairs terminate in their respective leg neuromeres. The remaining thin axon projects to the suboesophageal ganglion ipsilateral to the hair filled and terminates in the same region as a branch of the labellar dorsalmost axon. These data suggest that axonal arbors from multimodal hairs have a limited functional and somatotopic organization in the blowfly CNS.

Animals↗

PubMed2006

Multimodal prophylaxis for THA with mechanical compression.

We used mechanical thromboembolism prophylaxis using intraoperative thigh-calf pneumatic compression and other measures in 1032 consecutive primary and revision total hip arthroplasties. No chemical prophylactic measures were used until after duplex ultrasonography was performed by experienced technologists before discharge. Asymptomatic proximal thrombi were treated with low molecular weight heparin and warfarin, whereas those patients with a negative scan or distal thrombi only were advised to take aspirin 325 mg twice a day for 6 weeks. Regional anesthesia was used in 95% of the arthroplasties. Using this protocol, the 30-day mortality was 0.3%. There was one autopsy-proven fatal pulmonary embolism (0.09%). One other patient died suddenly with cardiac arrest after abdominal pain and vomiting, but no autopsy was performed. Symptomatic pulmonary embolism occurred in seven patients (0.7%), four occurring early and three late. Only one of these seven patients had a positive duplex scan. Deep vein thrombosis occurred in 41 patients (3.9%) and 35 remained asymptomatic. We observed no association between type of surgery (primary or revision), age, gender or preoperative diagnosis and pulmonary embolism or deep vein thrombosis. The data confirm the efficacy of a multimodal protocol with thigh-calf mechanical prophylaxis for almost all patients undergoing primary or revision total hip arthroplasty.

Adult↗

PubMed2004

Subthalamic nucleus stimulation for advanced Parkinson's disease: how to find a far medial STN.

In a patient with advanced Parkinson's disease, an anatomically deviant far medial subthalamic nucleus (STN) posed problems in the placement of DBS electrodes for chronic high frequency (HF) stimulation despite the use of multimodal targeting with 1) statistical atlas data, 2) T (2)-weighted (T (2)W) magnetic resonance imaging (MRI), 3) microelectrode recording, and 4) clinical testing with macro stimulation. Diagnostic T (2)W MRI suggested that the patient's STN was in a typical location and seemed to confirm the statistical atlas-based planning. Intraoperatively, cell activity recording (MER) with five parallel electrodes could not reveal any STN typical activity profile and electrical stimulation was not able to disclose a medial or lateral displacement of the electrodes. The operation was discontinued and postoperative stereotactic CT confirmed that the correct target area had been approached during the operation. Postoperative T (2)W MRI now disclosed a left STN which was 2 mm medial of the initial target and lead to a further medial target definition and finally to a successful DBS placement. In conclusion, finding a deep seated DBS target like the STN can be difficult in cases with an extremely deviant anatomy even if reiterative sophisticated multimodal planning is used. In the presented case we applied the integrated information from intraoperative MER, macrostimulation and postoperative imaging work-up and were able to complete DBS implantation successfully.

Brain Mapping↗

PubMed2003

Ten-year outcome after combined modality therapy for inflammatory breast cancer.

PURPOSE: To evaluate the long-term outcome of combined modality therapy for inflammatory breast cancer. METHODS AND MATERIALS: The data from 54 women treated between 1983 and 1996 for inflammatory breast cancer were analyzed. Patients with metastatic disease or disease progression on induction chemotherapy were excluded. Induction chemotherapy was given to 52 patients. Mastectomy was performed in 52 patients. Radiotherapy was delivered to the breast or chest wall and regional lymph nodes in all patients. The median follow-up for all patients was 5.1 years. RESULTS: The 5- and 10-year overall survival rate was 56% and 35%, respectively; the corresponding relapse-free survival rates were 49% and 34%. Patients with a pathologic complete response after chemotherapy with or without preoperative radiotherapy had better 5- and 10-year overall survival rates (65% and 46%, respectively) and 5- and 10-year relapse-free survival rates (59% and 50%, respectively) compared with patients without a pathologic complete response. Those patients had a 5- and 10-year relapse-free survival rate of 45% and 27%, respectively. Locoregional failure at 5 and 10 years was 8% and 19%, respectively. CONCLUSION: The outcomes for patients completing multimodality therapy compare favorably with published data; however, the exclusion of patients with progression during induction chemotherapy may account in part for these results. The pathologic complete response rate was found to be an important prognostic factor. Selected patients with inflammatory breast cancer have the potential for long-term survival.

Adenocarcinoma↗

PubMed2001

[Fusion of MRI, fMRI and intraoperative MRI data. Methods and clinical significance exemplified by neurosurgical interventions].

The aim of this work was to realize and clinically evaluate an image fusion platform for the integration of preoperative MRI and fMRI data into the intraoperative images of an interventional MRI system with a focus on neurosurgical procedures. A vertically open 0.5 T MRI scanner was equipped with a dedicated navigation system enabling the registration of additional imaging modalities (MRI, fMRI, CT) with the intraoperatively acquired data sets. These merged image data served as the basis for interventional planning and multimodal navigation. So far, the system has been used in 70 neurosurgical interventions (13 of which involved image data fusion--requiring 15 minutes extra time). The augmented navigation system is characterized by a higher frame rate and a higher image quality as compared to the system-integrated navigation based on continuously acquired (near) real time images. Patient movement and tissue shifts can be immediately detected by monitoring the morphological differences between both navigation scenes. The multimodal image fusion allowed a refined navigation planning especially for the resection of deeply seated brain lesions or pathologies close to eloquent areas. Augmented intraoperative orientation and instrument guidance improve the safety and accuracy of neurosurgical interventions.

Adult↗

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↗

PubMed2005

Functional recovery after open versus laparoscopic colonic resection: a randomized, blinded study.

BACKGROUND: Laparoscopic colonic surgery has been claimed to hasten recovery and reduce hospital stay compared with open operation. Recently, enforced multimodal rehabilitation (fast-track surgery) has improved recovery and reduced hospital stay in both laparoscopic and open colonic surgery. Since no comparative data between laparoscopic and open colonic resection with multimodal rehabilitation are available, the value of laparoscopy per se is unknown. METHODS: In a randomized, observer-and-patient, blinded trial, 60 patients (median age 75 years) underwent elective laparoscopic or open colonic resection with fast-track rehabilitation and planned discharge after 48 hours. Functional recovery was assessed in detail during the first postoperative month. RESULTS: Median postoperative hospital stay was 2 days in both groups, with early and similar recovery to normal activities as assessed by hours of mobilization per day, computerized monitoring of motor activity assessed, pulmonary function, cardiovascular response to treadmill exercise, pain, sleep quality, fatigue, and return to normal gastrointestinal function. There were no significant differences in postoperative morbidity, mortality, or readmissions, although 3 patients died in the open versus nil in the laparoscopic group. CONCLUSION: Functional recovery after colonic resection is rapid with a multimodal rehabilitation regimen and without differences between open and laparoscopic operation. Further large-scale studies are required on potential differences in serious morbidity and mortality.

Activities of Daily Living↗

PubMed1999

Comparison of dynamic and integrated light-scattering techniques in the study of the interaction of Candida rugosa lipase with DPPC liposomes.

Dynamic light-scattering (DLS) and wide angle integrated light-scattering (WAILS) spectroscopies were evaluated in the study of binding of Candida rugosa lipase (CRL) with 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) liposomes. The use of cumulants analysis on DLS data allowed for the determination of general lipase-liposome-binding trends. Particle intensity distributions obtained from DLS data by a discrete inversion method revealed the different populations created upon lipase-liposome interactions. Using a discrete inversion technique on WAILS data, not only these populations could be differentiated but also accurate number distributions were obtained in short periods of time. Both DLS and WAILS are excellent tools for the study of lipase binding to lipid vesicles; however, care must be exercised in the analysis of the experimental data whenever particle size distributions are multimodal. The selection of the light scattering technique will depend on the information required.

1,2-Dipalmitoylphosphatidylcholine↗

PubMed2003

Pathologic and molecular aspects of soft tissue sarcomas.

This article retains the conventional approach to the classification of soft tissue sarcomas, dividing them into several major histogenetic categories based on their overall microscopic appearance, tissue differentiation pattern, and biologic potential. The author advocates a multimodal approach, in which four distinctive data sets--clinical, radiographic, microscopic, and, in some cases, molecular--are considered to establish the diagnosis and treatment plan. Such step-wise analysis is more likely to lead to consistency and accuracy as compared with an intuitive approach based on fragmentary data. The author describes individual lesions of soft tissue as clinicopathologic entities and believes that they can be more accurately diagnosed and appropriately treated with the help of data generated by a multidisciplinary team. In addition, this article emphasizes the need to use emerging molecular techniques that can provide important clues for both diagnosis and prognosis.

Clinical Trials as Topic↗

PubMed2004

Nine months of multicomponent behavioral treatment for ADHD and effectiveness of MTA fading procedures.

We examined 9-month data from the 14-month NIMH Multimodal Treatment Study of Children with ADHD (the MTA) as a further check on the relative effect of medication (MedMgt) and behavioral treatment (Beh) for attention-deficit/hyperactivity disorder (ADHD) while Beh was still being delivered at greater intensity than at 14-month endpoint, and conversely as a check on the efficacy of the MTA behavioral generalization/maintenance procedures. Intention-to-treat analysis at 9 months showed essentially the same results as at 14 months, after Beh had been completely faded; MedMgt and the combination (Comb) of medication and Beh were significantly superior to Beh and community care (CC) for ADHD and oppositional-defiant (ODD) symptoms, with mixed results for social skills and internalizing symptoms. All treatment-group differences examined as changes in slopes from 9 to 14 months were nonsignificant (we found general improvement for all groups). Slopes from baseline to 9 months correlated highly (r > .74, p < .0001) with slopes from baseline to 14 months for all groups. The time function from baseline to 14 months showed a significant linear, but not quadratic, trend for the main outcome measure (a composite of parent- and teacher-rated ADHD and ODD symptoms) for all groups. Findings suggest that in contrast to the hypothesized deterioration in the relative benefit of Beh between 9 and 14 months (after completion of fading), the MTA Beh generalization and maintenance procedures implemented through 9 months apparently yielded continuing improvement through 14 months, with preservation of the relative position of Beh compared to other treatment strategies.

Analysis of Variance↗