PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Computational pathology”

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 19 recordsLinked to original sources
PubMed2026

Multimodal alignment improves generalizability of genomic biomarker prediction in computational pathology.

Computational pathology models that use digitized histopathology whole-slide images have the potential to become a cost-effective and scalable alternative to molecular assays for the prediction of genomic biomarkers, a key task in precision oncology. However, as new genomic biomarkers are discovered or quantified, large, labeled datasets must be prospectively collected to train new models. To address this challenge, we developed multimodal alignment for biomarker learning and generalization (MARBLE), a multimodal contrastive pretraining strategy that integrates structured biomarker knowledge into representation learning of histopathology images. MARBLE aligns histopathology-derived representations with representations of genomic biomarkers generated by a large language model (LLM) and a protein language model (PLM). This biologically informed alignment enables data-efficient generalization to novel, out-of-distribution biomarkers. Using the MSK-IMPACT cohort of over 40,000 patients across multiple biomarker panel versions, we design experiments grounded in real-world data to demonstrate the value of our proposed approach.

CP: computational biology↗
PubMed2025

Computational Pathology for Accurate Prediction of Breast Cancer Recurrence: Development and Validation of a Deep Learning-Based Tool.

Accurate recurrence risk stratification is crucial for optimizing treatment plans for breast cancer patients. Current prognostic tools like Oncotype DX offer valuable genomic insights into hormone receptor-positive and human epidermal growth factor receptor-negative patients but are limited by cost and accessibility, particularly in underserved populations. In this study, we present Deep-Breast-Cancer-Recurrence (BCR)-Auto, a deep learning-based computational pathology approach that predicts breast cancer recurrence risk from routine hematoxylin and eosin-stained whole slide images. Our methodology was validated on 2 independent cohorts: The Cancer Genome Atlas Program breast cancer data set and an in-house data set from The Ohio State University. Deep-BCR-Auto demonstrated robust performance in stratifying patients into low- and high-recurrence risk categories. On The Cancer Genome Atlas Program breast cancer data set, the model achieved an area under the receiver operating characteristic curve of 0.827, significantly outperforming the existing weakly supervised models (P = .041). In the independent The Ohio State University data set, Deep-BCR-Auto maintained strong generalizability, achieving an area under the receiver operating characteristic curve of 0.832, along with 82.0% accuracy, 85.0% specificity, and 67.7% sensitivity. These findings highlight the potential of computational pathology as a cost-effective alternative for recurrence risk assessment, broadening access to personalized treatment strategies. This study underscores the clinical utility of integrating deep learning-based computational pathology into routine pathological assessment for breast cancer prognosis across diverse clinical settings.

Humans↗
PubMed1978

Computed tomography in genitourinary pathology.

Computed tomography (CT) provides accurate localization of pathologic processes within the abdomen and also frequently contributes information regarding the nature of the process. A review of 50 patients with suspected genitourinary pathology studied with CT suggests that this new noninvasive diagnostic modality can provide increased diagnostic accuracy but that it will not replace conventional abdominal radiography and intravenous urography as the primary screening examination for such patients. Situations in which CT has proved valuable include: (1) evaluation of areas which are relatively "blind" on conventional radiography: anterior and posterior margins of kidneys, paravertebral and midline portions of retroperitoneum, and base of bladder; (2) evaluation of patients with severely compromised function of one or both kidneys, including assessment of amount of functioning renal parenchyma and possible demonstration of cause of obstructive hydronephrosis; (3) evaluation of patients in whom intravenous urography is contraindicated; (4) differentiation of solid from cystic masses of kidney; (5) evaluation of spread of tumor outside boundaries of kidney; and (6) staging of bladder tumors.

Abscess↗
PubMed1977

Computer-assisted pathology encoding and reporting system (CAPER).

An on-line computer-assisted pathology encoding and reportying system (CAPER) has been developed by the Department of Pathology and Laboratory of Computer Science of the Massachusetts General Hospital for a department of surgical pathology that processes more than 25,000 specimens yearly. CAPER performs clerical functions, including the accessioning of specimens, monitoring their state of completion, production of log books, billing, statistics, and transfer of diagnoses to other hospital departments. It also permits instantaneous display of all diagnoses rendered within two years, printout within 24 hours of all older diagnoses for any patient, and retrieval of all specimens with any given diagnosis, further defined by any data item (e.g., age) stored in the computer file.

Computers↗
PubMed1987

Rapid cytophotometry and its application to diagnostic pathology.

Computer processing of microscopic imagery of cervical lesions can provide novel quantitative diagnostic information. Interpretation of this information, from the level of the image itself to that of diagnostic truth, calls for the use of appropriate data structures. The resolution of the problems of diagnostic consensus and truth may be furthered by the use of these techniques in conjunction with diagnostic expert systems. The impact of our expanded ability to differentiate and to extract diagnostic and prognostic information will lead to changes in the way we practice clinical pathology. Technologic advances have now made it possible to perform quantitative measurements in a very rapid fashion as a practical part of routine clinical pathology activity.

Carcinoma↗
PubMed1988

Computed tomography: pathologic correlation in lung disease due to tocainide.

High resolution computed tomography (CT) was used to demonstrate the active lung phase of a patient with tocainide pulmonary toxicity. Gallium uptake in the lungs, an interstitial pneumonitis with alveolar macrophages on transbronchial biopsy, and reversal of process with discontinuation of the drug indicate a pulmonary drug reaction. The high resolution CT scan defined alveolar and interstitial components to the lung abnormality and correlated closely with the pathologic findings. High resolution CT is a useful method for demonstration of fine lung morphology.

Aged↗
PubMed2026

Beyond Morphology: Reframing Lymph-Node Metastasis Prediction Through Clonal Ecology-Decades-Long Genomic Instability and Polyclonal-to-Monoclonal Transitions as the Missing Dimension in Cancer.

Recent whole-genome, lineage-tracing, single-cell, and spatial studies have reshaped our understanding of tumor evolution, revealing that cancers can arise from polyclonal populations, undergo decades-long genomic instability before clinical detection, and progress through dynamic changes in subclonal composition, cellular state, and ecological organization. These findings challenge the assumption underlying morphology-based prediction models that metastatic risk can be inferred from static histological features alone. Here, we revisit lymph-node metastasis prediction in colorectal cancer through clonal ecology, integrating computational pathology with evolutionary oncology. Drawing on the subclonal switchboard model proposed in 2012 and subsequent artificial intelligence (AI)-enabled approaches for tracking dominant and dormant subclones, we synthesize evidence that metastatic potential reflects clonal ancestry, evolutionary timing, spatial niche architecture, cellular plasticity, intercellular interactions, dormancy, and treatment-driven shifts in subclonal fitness. We define five complementary methodological pillars for operationalizing clonal ecology: single-cell transcriptomics for resolving rare subclones, evolutionary trajectories, and adaptive cell states; lineage tracing and phylogenetics for reconstructing clonal ancestry and divergence; spatial transcriptomics and genomics for mapping subclonal geography and tumor-stromal-immune interactions; longitudinal liquid biopsy surveillance for monitoring residual disease, clonal turnover, and emerging resistance; and AI-enabled multimodal integration for connecting histopathology, genomics, spatial biology, and longitudinal data into predictive ecological-state models. Multiple-instance learning and pathology foundation models provide scalable computational foundations for evolution-aware prediction. Translationally, dormant subclones represent actionable reservoirs of recurrence. A longitudinal clinical and experimental study of KMT2A-rearranged acute myeloid leukemia further supports central predictions of the subclonal switchboard framework by demonstrating treatment-associated shifts in subclonal dominance, persistence of cryptic adaptive programs, and ecological rewiring during resistance and relapse. We propose clonal ecology as a measurable dimension for extending morphology-driven prediction toward integrative models that anticipate evolutionary transitions, identify therapeutic windows, and proactively constrain adaptive tumor ecosystems before resistant or metastatic subclones achieve clinical dominance.

Humans↗
PubMed1990

Computed tomography of metastatic cervical lymph nodes. A clinical, computed tomographic, pathologic correlative study.

A retrospective comparative study of 63 neck dissections was undertaken to evaluate further the accuracy of high-resolution computed tomography (CT) in the detection of nodal metastases, as previous studies have indicated a trend toward the superiority of CT scanning over palpation. The respective values of neck examination, CT scanning, and histopathologic examination were assessed in 51 patients with head and neck cancer who underwent a total of 63 neck dissections. The overall agreement between clinical examination findings and histopathologic findings was 92% vs 81% for CT scanning. A retrospective analysis of the CT findings failed to reveal greater accuracy. We found nodes measuring 10 mm or more with central low density always to be malignant. Because CT scanning seems to offer little advantage over palpation in the nonirradiated neck, it should not be regarded as an essential tool in the staging of nodal disease. After radiation therapy, as neck dissection is only performed because of clinical or radiologic suspicion, CT scanning is of utmost importance.

False Negative Reactions↗
PubMed1978

Radiation necrosis of the brain: correlation between computed tomography, pathology, and dose distribution.

Delayed necrosis of the brain following exposure to high doses of radiation is a recognized hazard. It is important to predict the nature of these postradiation changes, since surgical removal of radiation necrosis may reverse neurological deterioration of patients so affected. The present report stresses the importance of correlating the findings on computed tomography (CT) scans and zones of high radiation dose on dose reconstruction plans in order to differentiate between delayed radiation necrosis and tumors of the brain. This differentiation is not possible using the criteria of CT alone.

Adult↗
PubMed1991

[Approach to differential diagnosis of clinically related forms of chromosomal pathology using computers].

Multiple congenital developmental abnormalities account for a considerable share in the structure of the childhood morbidity, mortality and disability. Still, the differential diagnosis of the above abnormalities presents considerable difficulties because of the diversity of the forms and genetic pleomorphism. Using the method of rarefied templates of the "case--description term" type tried previously, a study was made of the possibility of differentiating between the clinically related forms of the chromosomal pathology 9p- and 11q- on the basis of phenotypic differences. The template was made up of 40 cases of 9p- and 40 cases of 11q-, accounting for 720 traits altogether. The "phenotypic nuclei"--traits occurring at a rate of over 25% were revealed for each syndrome and compared. Two approaches to the differentiation between these syndromes were used, which may turn out instrumental for diagnosing the clinically related forms of multiple congenital developmental abnormalities of the non-chromosomal genesis. The potentialities and difficulties of the computer-aided differential diagnosis are under discussion.

Abnormalities, Multiple↗
PubMed1988

Correlative investigations of craniospinal anatomy and pathology with computed tomography, magnetic resonance imaging and cryomicrotomy.

A correlative computed tomographic-anatomic method was applied to multiplanar anatomic studies. The method was particularly valuable for comparative anatomic investigations of complex regions of the central nervous system. The description of CT and MR anatomy in this thesis is based either on direct CT-anatomic correlation of the same specimen, or on indirect MR-anatomic correlation with cryosectional images from cadavers. In sagittal partial saturation MR images with short repetition times, the pituitary fossa in 41 normal volunteers appeared inhomogeneous. A postero-inferiorly located high intensity signal correlated with an intrasellar fat pad in sagittal cryosectional images. The height of the pituitary gland in 38 normal volunteers was usually less than 8 mm and its upper surface was flat or concave. The cavernous sinus anatomy was studied in coronal and axial MR planes in seven normal volunteers and 15 patients in correlation with cryosectional images. The intracavernous cranial nerves were best shown in the coronal plane, in partial saturation and inversion recovery sequences and displayed as foci of high signals intensity. MR signs of a parasellar mass included obliteration of intracavernous venous spaces, displacement of the intracavernous portion of the internal carotid artery and bulging of the lateral wall of the cavernous sinus. The topographic anatomy of the cervical neuroforamina was investigated in axial, sagittal, coronal and oblique planes in a correlative CT-anatomic investigation in 19 specimens. In four normal volunteers, the surface coil MR images of the cervical neuroforamina were correlated with cryosectional images. Surface coil MR images in a plane perpendicular to the cervical nerve roots allowed to determine their relationship to intraforaminal structures and the boundaries of the foramen. The dorsal and ventral nerve roots were demonstrated with both CT and MRI. They were located in the lower half of the foramen at and below the intervertebral disc space. The dorsal nerve roots were in contact with the superior articular process. The ventral nerve roots abutted the uncinate process and the inferior portion of the foramen. In a biomechanical cervical spinal trauma study, experimental fractures were studied with CT in axial and sagittal planes. Sagittal anatomic images showed the fractures and soft tissue injuries. Non-displaced and horizontal fractures were generally difficult to detect on axial CT scans alone. In two pathologic spine specimens CT scans in axial, sagittal and coronal planes were compared with reformatted scans and cryosectional images. Direct CT images revealed four cervical spinal fractures whereas the reformatted images only showed one displaced pedicle fracture.(ABSTRACT TRUNCATED AT 400 WORDS)

Animals↗
PubMed1987

Magnetic resonance imaging of the solitary hepatic mass: direct correlation with pathology and computed tomography.

Magnetic resonance images (MRI) of the liver were obtained using a combination of short time inversion recovery (STIR) and spin echo (SE) sequences. These were correlated with comparable tissue slices generated from resected specimens obtained at partial hepatectomy. All 10 cases appeared to have solitary masses on contrast enhanced computed tomography (CT). Histological examination revealed five primary tumours (two hepatocellular carcinomas, two haemangiomas and one cholangiocarcinoma) and five metastatic tumours. The STIR images demonstrated a high signal intensity in all areas of viable tumour involvement and reduced signal intensity in regions of confluent necrosis with superimposed haemorrhage or calcification. This sequence also demonstrated additional areas of high signal intensity adjacent to several lesions which were not visible on CT. Microscopy of these regions in the specimens demonstrated no tumour involvement or steatosis and their precise cause remains obscure. All the lesions demonstrated on the CT images were visible on MRI and no additional lesions were discovered on detailed microscopical examination of the specimens. Delineation of the extent of the cholangiocarcinoma was a problem with both techniques. MRI showed no major advantage over CT except for a higher contrast of the lesion compared with normal liver and also a better delineation of the tumour mass.

Adenoma, Bile Duct↗
PubMed1988

Computed tomography and pathologic correlations of thymic lesions.

Computed tomographic and pathologic correlations of the thymus gland were assessed in 69 patients. The sensitivity of computed tomography (CT) for undifferentiated thymic pathology is 87.1%; the specificity is 85.7%. The sensitivity of CT for neoplasm or mass is 97.1%, the specificity is 97.1%. The sensitivity of CT for lymphoid follicular hyperplasia (LFH) is 71.4%, the specificity is 97.6%. Therefore, a normal-sized thymus gland on CT does not exclude LFH. Completely preserved fat planes between thymic mass and adjacent structures on CT usually indicate a benign (noninvasive) neoplasm; completely absent fat planes usually indicate a malignant (invasive) neoplasm; partially preserved fat planes are indeterminate in assessing invasiveness. CT is also useful in showing recurrence or remnants of thymic tissue in patients who have had a previous thymectomy.

Adolescent↗
PubMed1986

[Computerized records in vascular pathology. V. Computer-assisted management of records in venous pathology].

Computerized processing of data from patients with venous disease concerns mainly those with chronic venous insufficiency and venous thromboembolic disease. It is of particular interest for epidemiologic studies of venous disease, explorations (functional, phlebographic) and proposed therapies. Progress in these 3 sectors requires computerized processing of data from case-reports and standardization of the noting of these data.

Computers↗
PubMed1986

[Computerized records in vascular pathology. II. Computer-assisted management of records in cerebrovascular pathology].

Computerized processing of data from patients with cerebro-vascular disease should improve interpretation of results of complementary examinations by allowing their comparison with arteriography and operative findings. It should also allow comparison of postoperative clinical and angiographic findings in order to analyse results obtained in an identical manner, and to provide a basis for discussion of therapeutic indications. Spontaneous evolution of atheromatous lesions can finally be compared with the course after surgery.

Cerebral Angiography↗