News | AI uncovers new clue to unexplained infertility: disrupted hormonal rhythms
A study presented at the 28th European Congress of Endocrinology suggests that infertility may depend not only on whether hormone levels are high or low, but on whether hormones change at the right time and in the right rhythm.
Researchers developed an artificial intelligence (AI)-enabled wearable skin-sensor patch that continuously tracks reproductive hormone fluctuations rather than measuring a single concentration at one time. The technology could help clinicians detect hidden reproductive dysfunction missed by conventional tests and improve the success of natural conception and assisted reproductive treatment.
The study was led by Dr Tinatin Kutchukhidze of the University of Oxford and New Anglia University.
About 15%–30% of people with unexplained infertility have no clear abnormality on routine testing. Men are generally assessed for hypogonadism with a single morning testosterone test, while women are assessed mainly through menstrual cycles and levels of luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol, and progesterone.
Hormones, however, are not static values. They are biological signals with clear circadian rhythms and dynamic fluctuations.
Conventional tests capture only one point in time and overlook the timing and coordination among hormones.
The team first studied 102 men aged 22–38. All had morning total testosterone within the normal range of 12–35 nmol/L, although some had infertility or symptoms associated with low androgen levels.
The AI wearable skin sensor recorded testosterone every 15 minutes for 4 days. Even with “normal” laboratory results, symptomatic men showed clear disruption of testosterone rhythms.
These hidden rhythm abnormalities were closely associated with lower sperm concentration and symptoms of androgen deficiency.
Dr Tinatin Kutchukhidze said this was the first noninvasive AI patch to continuously track androgen rhythms in real time over several days. “It has generally been assumed that normal morning testosterone is enough to rule out clinically significant androgen deficiency. Our study challenges that assumption.”
The team also developed an AI measure for women called ERI, or Endocrine Rhythm Integrity.
The study included 312 women aged 18–22 who reported regular menstrual cycles, including women with confirmed fertility and those with unexplained infertility. Researchers analyzed luteal-phase hormone changes, basal body temperature, heart rate, and sleep patterns.
Women with unexplained infertility had significantly lower ERI scores even when hormone levels were within normal ranges.
Lower ERI scores were also associated with a higher risk of embryo implantation failure.
The findings suggest that some women with apparently normal cycles and hormone results may have hidden endocrine rhythm disorders that affect fertility.
Unlike conventional tests, ERI does not analyze a single hormone value. It assesses whether reproductive hormones change at the right time, in the right pattern, and in coordination with one another, Dr Tinatin Kutchukhidze explained.
AI rhythm analysis was substantially better than conventional testing at identifying subclinical reproductive dysfunction.
The findings support an emerging view that many endocrine disorders in men and women may involve abnormal hormone synchronization and biological rhythms rather than abnormal hormone quantities.
The team next plans to test whether the tool can predict actual pregnancy outcomes across larger and more diverse populations and reproductive conditions.
Dr Tinatin Kutchukhidze said the team hopes to shift reproductive medicine from outcome-based diagnosis to rhythm-based prediction. “In the future, clinicians may be able to identify risk, personalize interventions, and improve outcomes before infertility actually occurs.”
She added that the technology could also be used in transgender healthcare. Current hormone therapy monitoring relies mainly on intermittent blood tests, which do not reflect true hormone dynamics. Continuous monitoring could support more precise, personalized long-term hormone management.
If validated, wearable hormonal chronodiagnostics could become a new standard in endocrinology and reproductive medicine.
News | AI uncovers new clue to unexplained infertility: disrupted hormonal rhythms
News | AI uncovers new clue to unexplained infertility: disrupted hormonal rhythms
A study presented at the 28th European Congress of Endocrinology suggests that infertility may depend not only on whether hormone levels are high or low, but on whether hormones change at the right time and in the right rhythm.
Researchers developed an artificial intelligence (AI)-enabled wearable skin-sensor patch that continuously tracks reproductive hormone fluctuations rather than measuring a single concentration at one time. The technology could help clinicians detect hidden reproductive dysfunction missed by conventional tests and improve the success of natural conception and assisted reproductive treatment.
The study was led by Dr Tinatin Kutchukhidze of the University of Oxford and New Anglia University.
About 15%–30% of people with unexplained infertility have no clear abnormality on routine testing. Men are generally assessed for hypogonadism with a single morning testosterone test, while women are assessed mainly through menstrual cycles and levels of luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol, and progesterone.
Hormones, however, are not static values. They are biological signals with clear circadian rhythms and dynamic fluctuations.
Conventional tests capture only one point in time and overlook the timing and coordination among hormones.
The team first studied 102 men aged 22–38. All had morning total testosterone within the normal range of 12–35 nmol/L, although some had infertility or symptoms associated with low androgen levels.
The AI wearable skin sensor recorded testosterone every 15 minutes for 4 days. Even with “normal” laboratory results, symptomatic men showed clear disruption of testosterone rhythms.
These hidden rhythm abnormalities were closely associated with lower sperm concentration and symptoms of androgen deficiency.
Dr Tinatin Kutchukhidze said this was the first noninvasive AI patch to continuously track androgen rhythms in real time over several days. “It has generally been assumed that normal morning testosterone is enough to rule out clinically significant androgen deficiency. Our study challenges that assumption.”
The team also developed an AI measure for women called ERI, or Endocrine Rhythm Integrity.
The study included 312 women aged 18–22 who reported regular menstrual cycles, including women with confirmed fertility and those with unexplained infertility. Researchers analyzed luteal-phase hormone changes, basal body temperature, heart rate, and sleep patterns.
Women with unexplained infertility had significantly lower ERI scores even when hormone levels were within normal ranges.
Lower ERI scores were also associated with a higher risk of embryo implantation failure.
The findings suggest that some women with apparently normal cycles and hormone results may have hidden endocrine rhythm disorders that affect fertility.
Unlike conventional tests, ERI does not analyze a single hormone value. It assesses whether reproductive hormones change at the right time, in the right pattern, and in coordination with one another, Dr Tinatin Kutchukhidze explained.
AI rhythm analysis was substantially better than conventional testing at identifying subclinical reproductive dysfunction.
The findings support an emerging view that many endocrine disorders in men and women may involve abnormal hormone synchronization and biological rhythms rather than abnormal hormone quantities.
The team next plans to test whether the tool can predict actual pregnancy outcomes across larger and more diverse populations and reproductive conditions.
Dr Tinatin Kutchukhidze said the team hopes to shift reproductive medicine from outcome-based diagnosis to rhythm-based prediction. “In the future, clinicians may be able to identify risk, personalize interventions, and improve outcomes before infertility actually occurs.”
She added that the technology could also be used in transgender healthcare. Current hormone therapy monitoring relies mainly on intermittent blood tests, which do not reflect true hormone dynamics. Continuous monitoring could support more precise, personalized long-term hormone management.
If validated, wearable hormonal chronodiagnostics could become a new standard in endocrinology and reproductive medicine.
Source:
Compiled from online sources