Int J Mol Sci. 2025 Jun 23;26(13):6024. doi: 10.3390/ijms26136024.

ABSTRACT

Rare genetic movement disorders usually manifest early in life with dystonia, parkinsonism, chorea, or a combination thereof. These are often associated with neurodevelopmental delay, intellectual disability, speech problems, retinal abnormalities, seizures, ataxia, spasticity, or systemic features. Due to their vast number and pheno-genotypic heterogeneity, the diagnosis of these disorders can be challenging. However, recognising their core motor phenomenology as well as clinical, laboratory, and neuroradiological clues can expedite appropriate diagnostic workup, molecular diagnosis, and adequate treatment. In this review, we outline diagnostic clues to rare movement disorders (RMDs), focusing on those that present mainly with dystonia, parkinsonism, or paroxysmal dyskinesia due to genetic causes. Additionally, we provide a decision tree approach linking clinical, genetic, and imaging testing. Finally, we highlight selected RMDs that should not be missed, as they possess established treatments that can hinder their progression, prevent irreversible or life-threatening sequelae and, in certain cases, lead to complete symptom remission.

PMID:40649801 | DOI:10.3390/ijms26136024

Sci Rep. 2025 Jul 8;15(1):24442. doi: 10.1038/s41598-025-09744-y.

ABSTRACT

Following a confirmed genetic diagnosis, rare disease patients and their families encounter significant challenges in accessing diagnostic information and support. Patients and non-specialists are increasingly expected to interpret and share test results; however, existing standards are primarily designed for specialists. These standards fail to address the needs of resource-limited populations where low genomic literacy hampers accurate dissemination of genetic results. This research introduces RareInsight, an open-source, interactive dashboard designed to enhance the accessibility, comprehension, and collaboration of genetic data among patients, caregivers, clinicians, and researchers. Developed using shinydashboard, RareInsight was evaluated using whole exome sequencing data from skeletal dysplasia patients. It allows users to input and view Variant Call Format files and includes a searchable ClinVar variant table with filtering options, providing access to multiple resources based on search terms. RareInsight aims to simplify the dissemination of complex genetic information beyond the clinical setting. This dashboard serves as a pilot study demonstrating the potential of patient-centered interactive dashboards for the rare disease community.

PMID:40628872 | DOI:10.1038/s41598-025-09744-y

Orphanet J Rare Dis. 2025 Jul 7;20(1):344. doi: 10.1186/s13023-025-03879-x.

ABSTRACT

BACKGROUND: The cumulative economic burden of rare diseases surpasses that of common conditions, yet patterns of healthcare resource utilization (HRU) across rare diseases remain poorly characterized. This study leverages multimodal data collected during clinical care and through surveys to provide an in-depth evaluation of HRU across the disease journey of individuals with rare genetic diseases. Individuals with a confirmed diagnosis of Kleefstra syndrome (KS; n = 40) or SLC6A1 epileptic encephalopathy (SLC6A1; n = 30) were recruited. Structured and unstructured data were abstracted from participants' medical records. Encounters per person-year of follow-up were calculated and compared pre- and post-diagnosis. Parents/guardians completed surveys assessing the impact of the participant's diagnosis on their care.

RESULTS: Records were available for a median of 6.4 years of follow-up from 268 unique healthcare facilities (median per patient = 4.5 facilities). Numbers of healthcare encounters were not significantly different 1 year pre- and post-diagnosis for either condition; however, the proportion of specialty encounters pre- and post-diagnosis varied significantly. Genetics encounters decreased for both conditions post-diagnosis. Cardiology, sleep medicine, and radiology encounters increased in KS post-diagnosis; conversely, audiology encounters decreased in KS post-diagnosis, and radiology encounters decreased in SLC6A1 post-diagnosis. Among specialty encounter types assessed, general practitioner (e.g. primary care, including pediatrics) encounters were the most common type for KS participants and the second-most common for SLC6A1 participants (after neurology encounters) both 1 year pre- and post-diagnosis. The number of both echocardiograms and electrocardiograms (ECG) significantly increased in KS 1 year post-diagnosis. 68% of survey respondents indicated that the participant's care changed post-diagnosis.

CONCLUSIONS: Though there was no significant difference in the number of encounters pre- and post-diagnosis, significant changes in types of HRU suggest that diagnosis leads to more appropriate care and treatment. Advocacy organizations, researchers, drug developers, payors, and policymakers should consider the value of an early diagnosis to improve long-term outcomes and quality of life for patients and invest in measures that will shorten the time to diagnosis accordingly.

PMID:40624551 | DOI:10.1186/s13023-025-03879-x

Cir Cir. 2025 Jul 4. doi: 10.24875/CIRU.24000236. Online ahead of print.

ABSTRACT

OBJECTIVE: Skeletal muscle ion channelopathies are a rare genetically inherited orphan disease. Due to the unique characteristics of the symptoms of the disease, misdiagnosis of patients leads to irreversible losses. This study aims to raise awareness on this issue.

METHODS: 35 patients with a definitive diagnosis of skeletal muscle ion channelopathy were included in the study. The diagnoses of all patients were confirmed by gene analysis. Demographic and clinical characteristics of the patients were examined. After a definitive diagnosis was made, mimic symptoms and misdiagnoses were evaluated separately.

RESULTS: It was determined that 30 of the patients included in the study had multiple different diagnoses until they got the correct diagnosis. It is thought that due to delayed diagnosis or misdiagnosis, patients experience physical and mental loss, are exposed to ineffective drugs, and their daily lives are adversely affected, as well as serious cost losses.

CONCLUSIONS: It is stated that the names of misdiagnoses for imitation symptoms have changed with aging, and drug treatments are applied for each diagnosis. It is stated that health authorities should pay attention to this situation to reduce this.

PMID:40614251 | DOI:10.24875/CIRU.24000236

Echocardiography. 2025 Jul;42(7):e70225. doi: 10.1111/echo.70225.

ABSTRACT

We present a case of a primary cardiac synovial sarcoma, surrounding the right coronary artery and leaded to accelerate the blood flow of the tricuspid valve orifice.

PMID:40614057 | DOI:10.1111/echo.70225

BMC Nephrol. 2025 Jul 3;26(1):346. doi: 10.1186/s12882-025-04271-4.

ABSTRACT

BACKGROUND: Estimating disease epidemiology using health insurance claims data is challenging due to the lack of specific diagnostic codes. This study demonstrates an approach to estimate the epidemiology of primary immunoglobulin A nephropathy (IgAN) using German Statutory Health Insurance (SHI) claims data.

METHODS: A retrospective observational study was conducted using data from January 1st, 2015 to December 31st, 2022. Two coding algorithms - one restrictive and one inclusive - were developed to identify primary IgAN cases based on ICD-10-GM codes. The restrictive algorithm identified cases based on histologically confirmed diagnoses, while the inclusive algorithm included a broader range of related diagnoses. Annual incidence and prevalence rates were calculated and extrapolated to the German population.

RESULTS: The mean prevalence rate from 2017-2022 ranged from 4.5 (restrictive codes) to 38.3 (inclusive codes) cases per 100,000 individuals, demonstrating the orphan nature of primary IgAN. Mean incidence rates ranged from 0.2 to 0.6 cases per 100,000 individuals. In 2022, this corresponded to an estimated 4,043 to 32,229 prevalent cases and 164 to 538 incident cases in Germany.

CONCLUSION: This study presents an approach for estimating disease epidemiology in the absence of specific diagnostic codes, using primary IgAN as an example. While our dual-algorithm method provides valuable insights into the incidence and prevalence of primary IgAN in Germany, it also highlights the need for consistent coding practices and specific diagnostic codes for accurate epidemiological assessments.

PMID:40610969 | DOI:10.1186/s12882-025-04271-4

Sci Rep. 2025 Jul 3;15(1):23780. doi: 10.1038/s41598-025-04428-z.

ABSTRACT

In the time of advancing medical technology, there is a critical issue concerning the misdiagnosis of diseases. The aim of this research is to significantly reduce the occurrence of misdiagnoses in medical practice by leveraging hypergraphs and genomic data to improve diagnostic accuracy. We have designed and implemented a sophisticated computational framework for phenotype-driven disease prediction that leverages hypergraphs and genomic data to enhance the accuracy of disease identification, leading to more precise and timely treatments for patients. The study employed robust ranking algorithms, on a sample of 2130 diseases, 4655 genes and 9541 phenotypes collected from reliable sources of Orphanet and Human Phenotype Ontology (HPO) database to achieve highly favorable outcomes. The proposed method outperforms existing state-of-the-art tools such as Phenomizer and GCN, in terms of both prediction accuracy and processing speed. Notably, it captures 50% of causal genes within the top 10 predictions and 85% within the top 100 predictions and the algorithm maintains a high accuracy rate of 98.09% for the top-ranked gene. In conclusion, our study demonstrated the effectiveness of robust ranking algorithms and hypergraph framework in achieving accurate and reliable results for disease diagnosis. While the study provides valuable insights, it is important to note its limitations, such as the sample size and scope of diseases considered. Future research could explore the integration of additional data sources and refinement of algorithms to further enhance diagnostic capabilities. Overall, this study underscores the potential of algorithmic hypergraph based approaches in advancing medical diagnostics and improving healthcare delivery.

PMID:40610497 | DOI:10.1038/s41598-025-04428-z

Epidemiol Prev. 2025 Mar-Jun;49(2-3):181-189. doi: 10.19191/EP25.2-3.A776.025.

ABSTRACT

OBJECTIVES: to provide the epidemiological framework of those affected by rare diseases resident in the Campania Region (Southern Italy), using the data entered in the Campania Region Rare Disease Registry, acquiring information potentially useful for regional planning.

DESIGN: observational retrospective cohort study on patients with rare diseases included in the Regione Campania Rare Disease Registry from 01.01.2022 to 31.12.2022.

SETTING AND PARTICIPANTS: population included in the Rare Disease Registry and resident in the Campania Region as at 31.12.2022.

MAIN OUTCOME MEASURES: using the data entered in the Regione Campania Rare Disease Registry, the cumulative incidence (I) of patients with rare diseases resident in Campania was calculated, stratified by age group and rare disease group with the respective 95% confidence intervals (IC95%). Standardised cumulative provincial incidences were also calculated. These are reported using a multiplication factor of 100,000.

RESULTS: the incidence of patients with rare diseases in the Campania Region is 50.0 (IC95% 49.4-50.6) per 100,000 inhabitants in the year 2022. Furthermore, the rare disease groups with the highest incidence per 100,000 inhabitants are diseases of the central and peripheral nervous system (I: 8.32 per 100,000 inhabitants) and congenital malformations, chromosomopathies and genetic syndromes (I: 8.52 per 100,000 inhabitants). Moreover, the age groups in which the incidence is highest are in the paediatric age group.

CONCLUSIONS: an epidemiological framework of the Campania Region on rare diseases such as this one for the year 2022 is fundamental for national and regional planning in order to improve the care and quality of life of people affected by rare diseases, who often feel neglected by society. Sharing this type of information also draws attention to the need for faster diagnosis and the specialisation of new centres.

PMID:40605728 | DOI:10.19191/EP25.2-3.A776.025

Nature. 2025 Jul;643(8070):47-59. doi: 10.1038/s41586-025-09096-7. Epub 2025 Jul 2.

ABSTRACT

From fertilization onwards, the cells of the human body acquire variations in their DNA sequence, known as somatic mutations. These postzygotic mutations arise from intrinsic errors in DNA replication and repair, as well as from exposure to mutagens. Somatic mutations have been implicated in some diseases, but a fundamental understanding of the frequency, type and patterns of mutations across healthy human tissues has been limited. This is primarily due to the small proportion of cells harbouring specific somatic variants within an individual, making them more challenging to detect than inherited variants. Here we describe the Somatic Mosaicism across Human Tissues Network, which aims to create a reference catalogue of somatic mutations and their clonal patterns across 19 different tissue sites from 150 non-diseased donors and develop new technologies and computational tools to detect somatic mutations and assess their phenotypic consequences, including clonal expansions. This strategy enables a comprehensive examination of the mutational landscape across the human body, and provides a comparison baseline for somatic mutation in diseases. This will lead to a deep understanding of somatic mutations and clonal expansions across the lifespan, as well as their roles in health, in ageing and, by comparison, in diseases.

PMID:40604182 | DOI:10.1038/s41586-025-09096-7

Comput Biol Med. 2025 Sep;196(Pt A):110610. doi: 10.1016/j.compbiomed.2025.110610. Epub 2025 Jul 1.

ABSTRACT

The use of artificial intelligence (AI) techniques is significantly changing the analysis of medical images, accelerating and standardizing the diagnosis process. To train an AI model, however, a large dataset is typically required, especially when using the most powerful techniques. Therefore, not all specialties are taking advantage of AI techniques in the same way. For instance, they are seldomly used in areas such as the diagnosis of rare diseases since, due to their low prevalence, not enough data are typically available to train an AI model. In this paper, we address the use of AI techniques to diagnose a particular rare disease: Collagen VI-related Congenital Muscular Dystrophy from confocal microscopy images. We apply both classical machine learning and modern deep learning techniques and we show that, when using the appropriate data management and training procedures, one can successfully derive a highly-accurate classifier even with a limited amount of training data. Due to the generality of the explored techniques, this conclusion is likely to hold also for most of the rare diseases whose diagnosis relies on the examination of histological images.

PMID:40602312 | DOI:10.1016/j.compbiomed.2025.110610

ACS Nano. 2025 Jul 2. doi: 10.1021/acsnano.5c05498. Online ahead of print.

ABSTRACT

Globoid cell leukodystrophy (GLD) is a rare hereditary inborn error of metabolism due to recessive mutations that cause loss of function of the enzyme galactosylceramidase (GALC). This results in the accumulation of the sphingolipids galactosylceramide (GalCer) and galactosylsphingosine (GalSph) in the lysosomes of neuronal cells. The accumulated GalCer and GalSph in cerebral macrophages of GLD patients are neurotoxic to oligodendrocytes and Schwann cells, leading to demyelination in the nervous system. The disease typically presents with infantile onset in the first six months of life and death by age 2. Here, we identified a supramolecular structure of GalCer and GalSph that may contribute to GLD pathology. Using biophysical assays commonly used for studying proteinaceous amyloids, e.g., amyloid-specific dyes, microscopical imaging, and a series of analytical methods (FTIR, PXRD, and SAXS), we demonstrate that both GalCer and GalSph can self-assemble in vitro into highly organized fibrils reminiscent of fibrils of amyloidogenic proteins. These fibrils exhibit significant cytotoxicity to both neuronal and oligodendroglial cells. Using an inhibitor of the GALC enzyme in cell culture to mimic the GLD pathophysiology, we could detect the accumulation of these fibrils in cells. We also observed that small molecules, which are bona fide inhibitors of proteinaceous amyloids, effectively mitigated the formation of the GalCer and GalSph fibrillar structures in vitro. Finally, the small molecule ameliorated the cytotoxic effects of the sphingolipid fibrils in SH-SY5Y cells, suggesting a potential avenue for therapeutic intervention in GLD orphan disease.

PMID:40603002 | DOI:10.1021/acsnano.5c05498

Am J Hematol. 2025 Jul 2. doi: 10.1002/ajh.27739. Online ahead of print.

NO ABSTRACT

PMID:40600615 | DOI:10.1002/ajh.27739

Orphanet J Rare Dis. 2025 Jul 1;20(1):321. doi: 10.1186/s13023-025-03838-6.

ABSTRACT

BACKGROUND: Rare diseases affect small populations but present unique challenges in access to healthcare and social support. The needs of patients and their caregivers extend beyond medical treatments, impacting various aspects of their lives. This study provides a narrative overview of these diverse needs experienced by patients and caregivers.

METHODS: A rapid literature review was conducted in PubMed and Embase, including studies assessing needs in rare diseases. Following Cochrane guidelines, two researchers screened 1.419 articles (74%) double-blinded, followed by a single researcher screening the remaining 509 articles (26%). Two researchers collaboratively extracted data into an extraction table. To validate and complement these findings, two stakeholder consultations were held with representatives from patient organisations, healthcare providers, the pharmaceutical industry, and policymakers.

RESULTS: A total of 272 articles were included in the review, and respectively 25 and 33 participants participated in the consultations. The identified needs were categorized into two levels: (i) patient needs, and (ii) caregiver needs, along with one overarching transversal need: (iii) information needs. Patient needs spanned health, healthcare, and social dimensions. Psychological, mental, and emotional health were frequently highlighted, but also autonomy emerged as a significant need. Healthcare needs included gaps in timely and accurate diagnoses, underscoring the need for more awareness among healthcare providers and appropriate diagnostic tools. Coordinated multidisciplinary care and accessibility to care and treatments were also identified as essential, yet lacking. Socially, patients experienced unmet needs in support networks, workplace inclusion, education, and financial stability. Caregivers' needs were related to physical and mental health, social connection, and financial support. Information needs, affecting both levels and even extending to healthcare providers, underscored the demand for more comprehensive, accessible information on rare diseases, treatment options, healthcare services, and available social support.

CONCLUSION: This study underscores the complex needs of persons living with rare diseases and their caregivers, advocating for a holistic approach in healthcare policy. Beyond medical interventions, addressing timely diagnosis, coordinated care, and psychological support are essential. Policymakers must consider these multifaceted needs to enhance patient outcomes and foster an inclusive, patient-centred healthcare framework.

PMID:40598354 | DOI:10.1186/s13023-025-03838-6

Orphanet J Rare Dis. 2025 Jul 1;20(1):331. doi: 10.1186/s13023-025-03862-6.

ABSTRACT

BACKGROUND: Melorheostosis is a rare skeletal and connective tissue disorder with the estimated prevalence of 1/1,100,000. Low prevalence of rare diseases (RDs) can lead to suboptimal knowledge and expertise among clinicians.

METHODS: The European Registries for Rare Endocrine and Bone Conditions (EuRREB) facilitates collection of a set of Core Data Elements and a specific dataset within the 'condition specific module' of the Core Registry platform. The Rare Bone Disease Action Group of the European Calcified Tissue Society (ECTS) collaborated with ERN BOND to develop a specific dataset for Melorheostosis.

RESULTS: An initial dataset was shortened to 44 unique variables. In January 2023, the Melorheostosis condition specific module was published and now consists of 18 patients from 2 countries. The median age of patients was 49 years old (range 23-82) and female to male ratio was 15:3 (83.3%). Family history of Melorheostosis was negative for all patients. The most affected bones were lower limbs in 12 cases (66.7%). Specifically, spine, feet and ribs were involved each in 2 cases (11%), skull and pelvis-in one patient each (5.5%). Two patients (11%) suffered from more than 1 lesion. Hyperostosis was present in 3 patients (16.7%), skeletal deformity-in 6 (33%), joint stiffness - in 11 (61%), asymmetry-in 16 (88.9%), joint limitation-in 12 (66.7%) patients. Swelling and muscle atrophy were reported in 1 case each (5.5%), vascular abnormalities-in 2 cases (11%), skin abnormality in 1 case (5.5%). Pain was present in 14 from 18 patients (77.8%). Genetic testing was performed in 5 patients (27.7%).

CONCLUSION: A condition specific module, for Melorheostosis, within an established registry has been developed. This will serve a useful resource to inform clinicians about this rare disease, and can support several healthcare initiatives such as guidelines creation and healthcare improvement strategies.

PMID:40598201 | DOI:10.1186/s13023-025-03862-6

Orphanet J Rare Dis. 2025 Jul 1;20(1):319. doi: 10.1186/s13023-024-03464-8.

ABSTRACT

BACKGROUND: Mucopolysaccharidosis II (MPS II) is a rare, life-limiting lysosomal storage disease caused by deficient iduronate-2-sulfatase activity. The current standard of care for MPS II is intravenous enzyme replacement therapy (ERT), which has been shown to improve somatic signs and symptoms and to increase life expectancy by approximately 12 years. This study reported on the somatic disease burden and clinical requirements of adult male patients in the Hunter Outcome Survey (ClinicalTrials.gov Identifier: NCT03292887).

RESULTS: Of the 373 patients in the analysis, 88 (23.6%) had cognitive impairment and 332 (89.0%) had received ERT. Almost half of all ERT-treated patients (47.0%) had undergone surgery in adulthood; the most common surgery was hernia repair (17.8% of patients). Over one-third (38.6%) reported hearing aid use. The median 6-min walk test distance for 151 treated patients was 436.0 m at the latest assessment after 18 years of age. Cardiovascular signs and symptoms were present in 71.6% (192/268) of patients and 27.3% (60/220) reported oxygen dependency after 18 years of age. Approximately half (50.9%) of ERT-treated patients experienced at least one serious adverse event in adulthood, with the most common being respiratory disorders. Intravenous ERT was well tolerated, with a rate of serious infusion-related reactions in adulthood of 0.03 per 10 patient-years.

CONCLUSIONS: Overall, adult patients with neuronopathic and non-neuronopathic MPS II had a high disease burden and requirement for surgeries, emphasizing the need to continue multidisciplinary management and regular assessments in adulthood. Further research into the differences in care needs of adult patients with MPS II is warranted. Trial registration NCT03292887 .

PMID:40598289 | DOI:10.1186/s13023-024-03464-8

Genome Med. 2025 Jul 1;17(1):72. doi: 10.1186/s13073-025-01494-w.

ABSTRACT

BACKGROUND: RNA sequencing (RNA-seq) is emerging as a valuable tool for identifying disease-causing RNA transcript aberrations that cannot be identified by DNA-based testing alone. Previous studies demonstrated some success in utilizing RNA-seq as a first-line test for rare inborn genetic conditions. However, DNA-based testing (increasingly, whole genome sequencing) remains the standard initial testing approach in clinical practice. The indications for RNA-seq after a patient has undergone DNA-based sequencing remain poorly defined, which hinders broad implementation and funding/reimbursement.

METHODS: In this study, we identified four specific and familiar clinical scenarios, and investigated in each the diagnostic utility of RNA-seq on clinically accessible tissues: (i) clarifying the impact of putative intronic or exonic splice variants (outside of the canonical splice sites), (ii) evaluating canonical splice site variants in patients with atypical phenotypes, (iii) defining the impact of an intragenic copy number variation on gene expression, and (iv) assessing variants within regulatory elements and genic untranslated regions.

RESULTS: These hypothesis-driven RNA-seq analyses confirmed a molecular diagnosis and pathomechanism for 45% of participants with a candidate variant, provided supportive evidence for a DNA finding for another 21%, and allowed us to exclude a candidate DNA variant for an additional 24%. We generated evidence that supports two novel Mendelian gene-disease associations (caused by variants in PPP1R2 and MED14) and several new disease mechanisms, including the following: (1) a splice isoform switch due to a non-coding variant in NFU1, (2) complete allele skew from a transcriptional start site variant in IDUA, and (3) evidence of a germline gene fusion of MAMLD1-BEND2. In contrast, RNA-seq in individuals with suspected rare inborn genetic conditions and negative whole genome sequencing yielded only a single new potential diagnostic finding.

CONCLUSIONS: In summary, RNA-seq had high diagnostic utility as an ancillary test across specific real-world clinical scenarios. The findings also underscore the ability of RNA-seq to reveal novel disease mechanisms relevant to diagnostics and treatment.

PMID:40597352 | DOI:10.1186/s13073-025-01494-w

Sci Rep. 2025 Jul 2;15(1):22678. doi: 10.1038/s41598-025-08866-7.

ABSTRACT

Caregivers of persons with rare diseases (RDs) face elevated stress levels, caregiver burden (CB), financial pressure, and decreased quality of life (QoL). Since the Polish Rare Diseases Plan for 2024-2025 does not address caregivers' psychosocial needs, this study aimed to assess the experiences of Polish parents of persons with RD. A self-administered, anonymous, computer-assisted online survey was conducted between March and August 2024 to examine the relationship between parenting a person with RD and caregivers' QoL, CB, and financial well-being. The survey included 942 Polish caregivers of individuals with RDs. The study demonstrated a statistically significant negative association between perceived CB and all dimensions of parents' QoL - physical health, psychological health, social relationships, and environment - indicating a broad decline in QoL as CB increases. Financial well-being emerged as a consistent positive predictor of QoL and was shown to buffer the negative effects of CB, underscoring its role as a critical resource for caregivers. Additionally, CB was associated with adverse experiences related to the diagnostic odyssey and its perceived consequences. Our findings highlight that long-term caregiving for individuals with RDs imposes substantial emotional, financial, and social burdens. To effectively address these challenges, Polish health policy must move beyond the biomedical model and adopt a comprehensive approach that integrates psychological, social, and financial support for RD families. Future research should explore targeted interventions that strengthen caregiver resources and reduce systemic barriers to support.

PMID:40595391 | DOI:10.1038/s41598-025-08866-7

Sci Rep. 2025 Jul 2;15(1):22732. doi: 10.1038/s41598-025-08829-y.

ABSTRACT

Although significant advances have been made in the early detection of many cancers, challenges remain in the early diagnosis of rare cancers, including Wilms tumor, Clear Cell Sarcoma of the Kidney, Neuroblastoma, Osteosarcoma, and Acute Myeloid Leukemia, perhaps due to their relative obscurity and scarce data compared to common cancers. Application of artificial intelligence or deep learning has shown promising results in disease diagnosis including in their ability to diagnose cancers and detect their tissue of origin. However, their ability to detect rare cancers is yet to be comprehensively assessed. This motivated us to develop, RareNet, which leverages transfer learning of an established deep learning model, namely, CancerNet, to classify rare cancers. The transfer learning framework of RareNet utilized DNA methylation data of various biopsied rare cancers to learn epigenetic signatures of rare cancers. RareNet achieved an overall accuracy (F1 score) of ~ 96%, outperforming other machine learning models including Random Forest, K Nearest Neighbors, Decision Tree Classifier, and Support Vector Classifier.

PMID:40594942 | DOI:10.1038/s41598-025-08829-y

Nat Commun. 2025 Jul 1;16(1):5528. doi: 10.1038/s41467-025-60577-9.

ABSTRACT

Previous foundation models for fundus images were pre-trained with limited disease categories and knowledge base. Here we introduce RetiZero, a vision-language model that incorporates knowledge from over 400 fundus diseases. The model is pre-trained on 341,896 fundus images with accompanying text descriptions gathered from diverse sources across multiple ethnicities and countries. RetiZero demonstrates exceptional performance across various downstream tasks including zero-shot disease recognition, image-to-image retrieval, clinical diagnosis assistance, few-shot fine-tuning, and cross-domain disease identification. In zero-shot scenarios, it achieves Top-5 accuracies of 0.843 for 15 diseases and 0.756 for 52 diseases, while for image-to-image retrieval, it scores 0.950 and 0.886 respectively. Notably, RetiZero's Top-3 zero-shot performance exceeds the average diagnostic accuracy of 19 ophthalmologists from Singapore, China, and the United States. The model particularly enhances clinicians' ability to diagnose rare fundus conditions, highlighting its potential value for integration into clinical settings where diverse eye diseases are encountered.

PMID:40592857 | DOI:10.1038/s41467-025-60577-9

Stud Health Technol Inform. 2025 Jun 26;328:372-376. doi: 10.3233/SHTI250741.

ABSTRACT

There are more than 6000 known Rare Diseases (RDs), which are often multi-systemic, and about 200 rare cancers (RC) which are the RD of oncology. The knowledge and understanding of rare diseases and rare cancers is limited because of their rarity. The European Health Data Space (EHDS) is a set of regulations, standards and guidelines that aim to empower individuals to access and control their personal health data. Electronic Health Records (EHRs) form the basis for primary data sharing and information exchange in the EHDS, instrumental for improving the quality and safety of patient care, while supporting research, innovation, and policy-making. Patient Summary is one of the primary health data categories mandated by the EHDS. This paper reviews essential data elements for RD and RC patient summaries. Starting from typical use cases for planned and unplanned care, this paper proposes data element extensions for patient summaries that address the needs of RD and RC patients, bridging primary and secondary use of data.

PMID:40588948 | DOI:10.3233/SHTI250741